FI80038C - FOERFARANDE FOER FRAMSTAELLNING AV 6 -BROMMI- OCH / ELLER 6,6-DIBROMY-PENICILLANSYRA-1,1-DIOXIDER OCH NYA 6-DIAZO-PENICILLANSYRA-1,1-DIOXIDDERIVAT. - Google Patents

Abstract

Preparation of penicillanic acid 1,1-dioxide-and derivatives thereof by debromination of 6-alpha-bromo- and/or 6,6-dibromopenicillanic acid 1,1-dioxides. The debromination is effected in a water-containing medium by reduction with zinc in association with an acid having a pKa-value measured in water less than 3.5.

Description

1 S O O 3 8

Process for the preparation of 6α-bromo and / or 6,6-dibromo-penicillanic acid 1,1-dioxides and novel 6-diazopenicillanic acid 1,1-dioxide derivatives

The invention relates to a process for the preparation of 6α-bromo and / or 6,6-dibromo-penicillanic acid 1,1-dioxides of general formulas.

Br HS, CHaX

:: / \ /

H - C - CC I

I I IX

C - N - CH CH3 /

0 COOR

and / or 0 0 w

Br H S, CH 2 X.

• · / \ /

Br - C - t C, II

I I I \, C - N - CH 'CH3 /

O COOR

wherein X is hydrogen, a chlorine atom or an acetoxy radical and R

is hydrogen or a metal or ester radical.

These compounds are valuable because they can be debromated to give penicillanic acid 1,1-dioxide and its derivatives, which are useful β-lactamase inhibitors. The demodulation step for converting the compounds of this invention to β-lactamase inhibitors is disclosed in FI patent application 831301.

According to the invention, the compounds of the formulas I and II are prepared by diazotizing a 6β-aminopenicillanic acid 1,1-dioxide derivative of the general formula 2 80038 o o \ /

H HS CHaX

! : / \ /

H2N - C - C C III

I I l \ c - N - CH CH3 /

O 'COOR

wherein X and R are as defined above, followed by bromination of the diazotized compound with at least an equimolar amount of nitrating agent in the presence of 1-5 equivalents of strong inorganic or organic acid in a solution or suspension of a mixture of water and a partially or completely water-miscible organic solvent medium, wherein the amount of water present is from 1 to 20% by volume of the mixture containing at least equimolar amounts of hydrogen bromide and bromine and optionally an excipient, or at least equimolar amounts of alkyl nitrite or cycloalkyl nitrite as a solution or suspension in a non-hydroxyl-containing, inert, organic solvent medium added water or up to 10% by weight of water to form 6-diazopenicillanic acid-1,1-dioxide derivative of the general formula

0 P

V

H S CHaX

....... '</ \ /

N - N C - C C X

I I I \ C - N - CH 'CH3 /

O 'CO-OR

II

3 80038 wherein X and R are as defined above, optionally isolating said diazo compound from the reaction mixture as such, or - when R is hydrogen - as a salt, converting the diazo compound to the general formula I and / or II with 6α-bromo and / or 6,6-dibromopenicillanic acid To a 1,1-dioxide derivative or an ester thereof by adding a complex or mixture containing at least an equimolar amount of bromine and bromine-hydrogen and preferably also a tertiary amine.

If it is desired to obtain salts or esters of the compounds I and / or II, either the salt or ester of the starting compound III or the acids 1 and / or II are converted into salts or esters after they have been prepared according to the invention. Details of the reaction conditions in the diazo-action and bromination steps of the process of the invention are set forth below.

It is known from the description of NL patent application 8 001 285 (corresponding to GB patent application 2 045 755A) to prepare 6-α-bromo-penicillanic acid of the formula:

Br H S .CH3! ! / \ /

H - C - CC IV

I I IX

C - N - CH CH3 /

0 COOH

diazotizing 6-O-aminopenicillanic acid of formula: H HS CH3; ! / \ /

H2N - C - CC V

I I l \ C - N - CH CH3 /

O s COOH

(hereinafter referred to as 6-APA for brevity) with an alkali metal nitrite, for example sodium nitrite, in the presence of a large excess of sulfuric acid and a bromide anion donor, for example 4,80038 sodium bromide.

It should be noted that the diazotization-bromination process of Cignarella et al. Provides, as one skilled in the art can expect, relatively small amounts of 6,6-dibromo-penicillanic acid of the formula:

Br H

! i / S \ p x "CH3

Br C - - C

I I I yr

I I I 'CH- VI

C ------ N- CH 3 /

0 'COOH

a fact confirmed by the experiments carried out by J.P. Clayton /J.Chem.Soc. (C), 2123 (1969] / ·

The use of the Cignarella method also provides, based on experience, 6-hydroxy-penicillanic acid and one of the two possible isomers of 6-bromo-6-hydroxy-penicillanic acid having the formulas:

OH H OH H

! CH3 J / CH3 H - C -1 'c; Br ~~ c --- C c,

| i I '' 'CH Da III' 'CH

C— N -CH 3 C —N —'C «3 S \ rf

o 'ΛΛΤΤ ϋ rooH

COOH

VII VIII

whose formation of compounds was difficult to minimize in experiments to improve the diazotization bromination process of Cignarella, as well as to the considerable amount of degradation products. It is possible to distinguish almost selectively useful reaction products, i. all monobromides of formula IV and most of the dibromides of formula VI, by reaction in the presence of a water-immiscible solvent, preferably dichloromethane, followed by extraction with dichloromethane, etc. However, it has been found that the Cignarella method, although more or less useful in small laboratory scale, is not , suitable for manufacture on an industrial scale for the following reasons. First, due to the poor solubility of 6-APA, the reaction must be performed at a very high dilution rate. Second, the reaction generates so much heat that even on a scale using only about 0.5 moles of 6-APA, cooling with ice-salt mixtures is only sufficient to control the temperature while ensuring rapid reaction and stability of the compounds of formulas IV and VI at low pH. in the presence of copious amounts of water. Thirdly, the already mentioned instability of useful products at a pH of about 0.5 prevents long reaction times and extraction treatments, as well as mixing times, which are necessary when carrying out the preparation on an industrial scale. This effect was already noticeable on a laboratory scale when moving from 0.1 mole to 1 mole.

Dutch patent application 80 01285 also describes a variant of another method of diazotization bromination of 6-APA published by Clayton / J.Chem.Soc. (C), 2123 (1969)} is hereinafter referred to as the Clayton method / using bromine instead of sodium bromide.

The original Clayton process, which produces mainly dibromide of formula VI, can in practice provide a larger amount of useful product than the Cignarella process. This process also involves very significant heat generation and involves the inconvenient simultaneous addition of bromine and nitrite to a highly acidic solution of 6-aminopenicillanic acid.

The variant of the Clayton method used in Dutch patent application 80 01285 involves the slow portionwise addition of 6-aminopenicillanic acid to a mixture of bromine, sodium nitrite, sulfuric acid, dichloromethane and water. However, the use of this conversion on a large scale is questionable given the considerable amount of 6,80038 of heat that develops during the reaction and the very precise dosing of 6-amino-penicillanic acid required.

The compounds prepared by the process of the present invention are valuable intermediates for the synthesis of penicillanic acid 1,1-dioxide compounds of the formula:

Os yO

H% -S

S / CH2X

H ~ C-C

i i i 'soh ix

C-N-CH

/

O 'COOH

wherein X is as defined above, and for the preparation of their salts and esters (if X = H is called PAS).

These compounds have useful pharmacological properties, for example, as potent inhibitors of several species of β-lactamases present in various bacteria.

Compounds of formula III in which X = H are already known from European patent application 78300860.0 (Publication No. 0002927). The production methods disclosed in this application are carried out by oxidizing 6-APA in a conventional manner, comprising double protection, namely protection of both the amino group and the carboxy group. Although protection of the carboxy group may not in fact be necessary, protection of the amino group must be carried out before the oxidation reaction is carried out. This means two additional separate steps, i. protection and removal. 6-3-Amino-penicillanic acid 1,1-dioxide and its esters prepared in this way are hardly preferred starting materials for the desired synthesis of compounds of formula IX, from which an economical, large-scale industrial process can be deduced. Fortunately, in a series of patent applications, for example, European patent application 80201192.4 (publication number 0030771) describes a simple method for β-β-amino-

II

7 80038 for the synthetic preparation of penicillanic acid 1,1-dioxide and its esters by conventional methods starting from 1,1-dioxins of penicillins which can be prepared by a large-scale fermentation process such as benzylpenicillin, the process comprising a reaction in one vessel, for example benzyllipenicillin Preparation of 1,1-dioxide followed by in situ protective silylation of the carboxy group, formation of the imide chloride with phosphorus pentachloride and conversion of the imide chloride, preferably with isobutanol, to the iminoether, followed by hydrolysis to give the desired starting material.

Compounds of general formula IX in which X is a halogen atom are known from Dutch patent publication 8100209.

A compound of formula III in which X = H and R = H (sometimes abbreviated hereinafter as "amino-PAS"), when thus prepared in two steps from benzylpenicillin, is in principle a possible and economically advantageous starting material because in the first step benzylpenicillin -Cillin-1,1-dioxide is obtained on a large scale in 70-90% yield by simple and direct oxidation with permanganate.

Whether it is preferred to use, for example, amino-PAS (or a derivative thereof in which X is not H but a substituent) and its esters as economically preferred starting materials depends on whether or not these starting materials can be converted into compounds of formula I and / or II in preferred yields. In this connection, it has of course been necessary to develop suitable conditions for the diazotization bromination of 6-8-aminopenicillanic acid 1,1-dioxide.

It has been found that diazotization of a compound in the presence of amino-PAS bromine using the Clayton method produces very small amounts of di-bromide, mainly of formula II (X = H). Gradual replacement of water with acetonitrile 8 80038 produced a small increase, mainly in dibromide, but yields remained too low. Diazotization at low pH in the presence of bromine, as well as in the presence of bromide anions, occasionally yielded estimated actual yields, with a maximum of 40-45% of useful products. However, such yields were not obtained systematically and the products obtained were usually of considerable impurity.

As a result of extensive research and experimentation, a method for diazotizing a 6-g-amino-penicillanic acid 1,1-dioxide derivative of the general formula has now been invented.

0 O

H H \ / y

! ! β. / CH 2 X

H-N - C -t XC

Δ. ^ ^ 1 I ^ h3 · m

c - N-CH J

CO 'COOR

wherein X and R are as defined above, followed by bromination of the diazotized compound at a temperature between -20 and 30 ° C, preferably between -10 and 15 ° C, using at least an equimolar amount of nitrating agent in the presence of 1-5 equivalents of strong inorganic or organic acid, in a solution or suspension comprising a mixture of water and a partially or completely water-miscible organic solvent medium, the amount of water present being from 1 to 20% by volume, containing at least equimolar amounts of hydrogen bromine or bromine. If desired, an additive which promotes the bromination of the diazotized compound may also be used in amounts ranging from 10% to at least an equimolar amount of the carbon dioxide starting material.

Various organic compounds can be used as excipients, which can optionally be added to the reaction mixture, such as generally tertiary amines, pyridine N-oxides, sulfones and sulfoxides, and generally amides, ureas and imides, all of which have a nitrogen-hydrogen moiety next to the acyl moiety. various compounds, i.e. 9,80038 having an imine (C = NH) type structural unit, such as 1,1,3 / 3-tetraalkyl guanidines, as well as various non-hydroxyl derivatives of malonic acid. It should be noted, however, that other suitable additives may be used which are not specifically mentioned in this specification.

The excipients that may be used in the diazotization treatment of this part of the invention belong to six groups of organic compounds. They are generally tertiary amines, pyridine N-oxides, sulfones and sulfoxides, and compounds having an acylamino group in the chain or ring, at least one nitrogen-hydrogen bond (N-H), and compounds having a C = N-H moiety. In these six groups of excipients, the individual structure is of little importance, with the main limitation being that the excipient should be relatively inert to the bromination conditions used. As a result, most enamines, and in general compounds with highly bromine-reactive ethylene or alkylene units, are excluded from this group. Likewise, Ν, Ν-dialkyl aromatic amines whose aromatic ring is brominated very easily are generally not useful. Another notable limitation of the possible nitration of an excipient susceptible in this respect, which in principle results in a separable nitroso derivative, is chemically detrimental in its consequences because the amino group of 6-3-aminopenicillanic acid 1,1-dioxide is much more reactive to nitration.

It should be noted that if it is preferred in the description of the invention to add a nitrating agent to a mixture containing a compound of formula II, it is generally equally possible to add a compound of formula II in portions to a mixture containing a nitrating agent. The fact that this latter inverse method is generally not advantageous is due to health considerations, i. a significantly increased risk of permanent or temporary conversion of certain excipients to nitroso derivatives, such as N-nitrosamides or imides, which may or may not have a beneficial effect as nitrating agents ίο 80038 on nitration of a compound of formula III.

The reaction conditions of the process of the present invention for converting 6-6-amino-penicillanic acid 1,1-dioxide in strongly acidic solution to the compounds of formula I and II can be described as follows.

The compound of formula HI is added with stirring and continuous cooling with ice water to a mixture of a suitable organic solvent medium, 1-20% by volume of water, preferably 3-15% by volume and 1-5 acid equivalents of a strong inorganic or organic acid, preferably arylsulfonic acid, such as p-tolylsulfonic acid, sulfuric acid or hydrobromic acid. Hydrobromic acid and sulfuric acid are, of course, preferably added as aqueous solutions, but the water in such solutions, as well as the much smaller amount of water present in the starting material and commercially available arylsulfonic acids, which are usually sold as hydrates, is preferred for the reaction. The concentration of the compound of formula III in the solutions thus prepared is at least 5% and may be increased to about 10 overnight by appropriate selection of conditions.

At least an equimolar amount (relative to the amount of amino PAS used) of hydrogen bromide, optionally an excipient, and at least an equimolar amount of bromine are then added. These substances can be derived separately or, if possible, as a 1: 1: 1 mixture (e.g. pyridine hydrobromide perbromide) or hydrogen bromide and a complex of excipient and bromine, or a salt of bromine and excipient and hydrogen bromide. When performing a separate or partially separate addition, it is preferred to add the bromine or bromine-excipient complex last. It should be noted that when hydrobromic acid is used to dissolve the starting material and then release the nitrosol species from the added nitrating agent, and if the excipient is not added complex with hydrogen bromide at the same time, the total calculated amount of hydrogen bromide can be added in one batch to prepare the starting solution. As soon as complete or almost complete dissolution is achieved, at least an equimolar amount of nitrating agent is added at the selected heat 11,80038 state or temperature range. The temperature may vary between -15 and + 25 ° C, preferably between -5 and +10, and even more preferably between 0 and + 10 ° C. The nitrating agent may be a commercially available cheap alkali metal nitrite, for example sodium nitrite, which is added gradually or in batches in approximately equal batches. It is possible, although not generally preferred, to add a dropwise saturated aqueous solution, for example sodium nitrite, taking into account the amount of water in this solution. As the nitrating agent, commercially available, preferably inexpensive cycloalkyl nitrite or alkyl nitrite such as ethyl nitrite, 1-pentyl nitrite, 2-pentyl nitrite, etc., which is added dropwise, can also be used. The rate of addition of the nitrating agent is adjusted to maintain the selected temperature range, but the time used is usually limited to 10-30 minutes. The reaction is complete after stirring for about another 15-45 minutes at about the same temperature. The final reaction mixture may be treated differently depending on the particular conditions selected in which the reaction is carried out and the purposes which have led to the choice of these particular conditions.

By suitable organic solvent medium is meant a simple organic solvent or a mixture of at least two organic solvents which, under the conditions used, do not substantially affect the reaction and which may always contain 10% by volume or more of water. Unsuitable simple solvents include ketones because they affect diazotized products, as well as various alcohols. Preferred simple solvents are therefore e.g. ethyl acetate, methyl acetate, nitromethane, dioxane, tetrahydrofuran, 1,2-dimethoxyethane and acetonitrile, all of which have a relatively low boiling point and can be easily removed at the end of the reaction in one way or another. In general, the most preferred solvent is acetonitrile and the second best is 1,2-dimethoxyethane. Depending on the reaction conditions, one of the other solvents mentioned may be a suitable choice, but acetonitrile and immediately afterwards 1,2-dimethoxyethane are almost always as suitable as any other solvent. Although 1,2-12,80038 dimethoxyethane, and especially acetonitrile, can be removed and largely recovered rapidly by concentrating the final reaction mixture in vacuo, these very good solvents can be detrimental to health, and environmental purity regulations have become increasingly demanding in some countries. As a result, it has been investigated whether these good solvents can generally be replaced by a combination containing at least 80% by volume of a common, less harmful solvent which can be easily recovered by separating the layers during the separation treatment, such as dichloromethane and especially ethyl acetate. a dipolar aprotic solvent, which increases the solubility of the water and salts, but on the other side which often can not be recovered easily. Good solvents of this type are 1,1,3,3-tetramethylurea, but especially dimethyl sulfoxide, tetrahydrothiophene-1,1-dioxide (sulfolane) and a group of non-hydroxyl solvents whose structure is derived from malonic acid, in which case the carboxyl radicals are esterified with lower alkyl groups or replaced with a cyano group, for example dicyanomethane, ethyl cyanoacetate, diethyl malonate, which in addition to the three types of solvent act as excipients. Less suitable appear to be the most common fully N-alkylated, amide-type solvents such as dimethylacetamide and N-methylpyrrolidin-2-one.

The relative amount of water in the diazotization-bromine mixture is usually between 1 and 20% by volume (preferably 3-15%).

It is impossible to further define the optimal relative amount of water, as this varies from case to case and depends on a number of selected reaction conditions, such as the nature of the organic solvent medium, the nature of the strong acid used, the nature of the nitrating agent and any excipient used.

Although other strong acids, i.e. strong organic acids, can sometimes be used to dissolve the compound amino-PAS or a derivative thereof and to obtain a sufficiently acidic solution.

II

In general, the most suitable acids are arylsulfonic acids, sulfuric acid and hydrobromic acid, and especially the latter acid. In this context, it should be noted that phosphoric acid is not well suited for use. The amount of 1 to 5 acid equivalents is mentioned above, but relatively good results are usually obtained using about 1.5 or more acid equivalents.

The variations in the diazotization-bromination series, which include the intermediate diazonium derivative, do not allow a more accurate general determination of the amount of strong acid required, because otherwise identical reactions of cycloalkyl nitrites and alkyl nitrite require more -riumnitriittiä.

According to an embodiment of the process according to the invention, it has been found that the yields have increased with the use of an excipient in certain cases.

One aspect of the diazo-functional bromination treatment of 6-β-amino-penicillanic acid-1,1-dioxide under conditions involving the formation of a diazonium intermediate is that, using the same general procedure but differently, the dibromide of formula II and the monobromide of formula I, can be prepared by modifying the ratios of hydrobromic acid bromine starting material of formula III to any excipient used. However, given the use of this method for the synthetic preparation of PAS, the following examples do not allow for a reliable and efficient classification of the excipients because the yield does not simply depend on the quality of the excipient. Each excipient requires the best conditions for its use, so that an absolute difference of 20% in actual yield between the two excipients, for example 70% versus 50%, obtained under otherwise identical or different conditions, does not necessarily, that the same difference remains after the most appropriate conditions have been determined for each of these excipients. It may be larger or smaller, as such a large difference is apparently an indication that the former is more effective, but not how much more effective. In the industrial scale process of the invention, the choice of the most suitable excipient depends not only on the yield and the price of the excipient, but also on ecological factors and ease of handling, making excipients which are not extractable or easily eliminated by washing organic solvent extracts such as tertiary amines, pyridine-1 oxides, hydantoin and succinic acid imide, or which can be completely recovered by extraction at neutral pH, such as caprolactam.

In view of the economic utility of the present invention, the excipients which may be used in the process of the invention are those which, for example in acetonitrile and in combination with about 50% excess bromine, provide compounds of formula I and / or II so that the estimated the actual yield is at least 65%. Suitable excipients which may be used in the specific situations defined by the invention include: (a) tertiary (cyclo) aliphatic amines having three groups attached to the nitrogen atom selected from straight or branched chain alkyl groups having no more than 6 (5-) tertiary saturated 5-7 membered heterocyclic amines in which the ring nitrogen atom is further attached to the (cyclo) alkyl group as defined above, while the ring may contain one oxygen atom or one oxygen atom; an additional nitrogen atom having a (cyclo) alkyl group as defined above, for example, N-ethylpyrrolidine, N-ethylpiperidine, N-methylmorpholine and N, N'-dimethylpiperazine;

II

80038 (c) tertiary, fully unsaturated heterocyclic amines including pyridine, quinoline, isoquinoline, pyramidine, pyrazine, N- (cyclo) alkylimidazole, N-vinylimidazole and (cyclo) alkyl groups as defined above; (d) pyridine-1-oxides; (e) sulfones and sulfoxides having two optionally substituted groups selected from the group consisting of alkyl, cycloalkyl, arylmethyl, arylethyl, aryl and trimethylene, tetramethylene and pentamethylene, and arylphenyl, biphenyl, 1- or 2 -naphthyl and pyridyl, and the aforementioned (cyclo) alkyl groups such as dimethyl sulfoxide, sulfolane, dimethyl sulfone and diphenyl sulfoxide; (f) alkyl, cycloalkyl, arylmethyl, arylethyl and arylcarbonamides, as well as N-mono (cyclo) alkyl and its N-monoaryl derivatives, and (cyclo) alkyl and aryl groups as defined above ", (g) saturated monocyclic 5-7 membered lactams such as caprolactam, (h) alkyl, (cyclo) alkyl, arylmethyl, arylethyl and arylsulfonamides, as well as their N-mono (cyclo) alkyl and N-monoaryl derivatives, and (cyclo) alkyl and aryl groups as defined above, (i) N-monoacyl ureides of general formula Q-CO-NH-CO-NH-Q1, wherein Q is (cyclo) alkyl, arylmethyl, arylethyl - or an aryl group, wherein the aryl group and the (cyclo) alkyl group are as defined above and are hydrogen or (cyclo) alkyl as defined above, and Q and Q1 may be bonded to each other to form a saturated 5- or 6-membered ring of methylene or in an ethylene unit optionally substituted by two methyl groups or one phenyl group such as hydantoin-5,5-dimethylhydantoin, 5-phenylhydantoin or N-phenylacetylurea; (j) imides of the type Q2-CO-NH-CO-Q3, wherein Q2 and 3Q are the same or different (cyclo) alkyl, arylmethyl, arylethyl or aryl groups, wherein the aryl and (cyclo) -1680 038 2 3 alkyl is as defined above and wherein Q and Q may also be bonded to each other to form a saturated 5- or 6-membered ring optionally substituted on the ethylene or propylene unit with one or two methyl groups or one phenyl group, for example N -benzoyl-benzenecarbonamide or succinimide; (k) disulfonylamines of the aryl-SC 1 -NH-SC 4 -aryl type, wherein the aryl may be the same or different and is as defined above; (l) recombinant imides of the type aryl-SO-NH-CO-Q 4, λ wherein Q is aryl or (cyclo) alkyl, wherein the aryl groups and the (cyclo) alkyl groups are as defined above; (m) phthalimide and saccharin; (n) guanidine and guanidines each having four lower alkyl groups and having a moiety C = N-H, wherein lower alkyl is as defined above, for example 1,1,3,3-tetramethylguanidine; (o) non-hydroxyl derivatives of malonic acid in which the carboxyl groups have been replaced by two cyano and / or N, N-di- (lower) alkylcarbamoyl and / or (lower) alkoxycarbonyl groups and the methylene group of which is optionally substituted by a lower alkyl group , and (p) α, ω-dicyanoalkanes, for example 1,2-dicyanoethane.

With respect to the nitrating agent, it has been found that although equimolar amounts of the nitrating agent greatly promote the reaction in the desired direction, maximum yields are usually obtained with an excess of about 15-40%. A significantly higher excess may result in lower yields, and the excess nitrating agent should not be greater than the amount of bromine used.

The resulting reaction mixtures are treated with an excess of bisulfite, for example, concentrated aqueous sodium metabisulfite to decompose any oxidizing agents still present, such as bromine, followed by the addition of dilute alkali metal.

II

17 80038 hydroxide therefore until the desired pH is reached. The further processing of the reaction mixture now diluted with water depends on the original purposes which have determined the choice of reaction conditions for the reactions carried out. In general, there are four possibilities for further processing, namely processing a, b, c and d.

Processing a.

If selective preparation is desired, the mixture is adjusted to pH 6-7 by appropriate selection of the reaction conditions, either dibromide of formula II or monobromide of formula I, followed by removal of the low-boiling organic solvent in vacuo, purification of the solution by extraction with water, preferably with dichloromethane and then extracting the desired product with, for example, ethyl acetate at pH 2-4. After the usual work-up, 6,6-dibromo-penicillanic acid 1,1-dioxide or 6-α-bromo-penicillanic acid 1,1-dioxide is obtained in substantially pure form, in the case of dibromide in a yield of 65% or more and in the case of monobromide of at least In 50% yield.

However, if these bromides prepared according to the invention are intended only as intermediates for the preparation of PAS, the treatment may be limited to the preparation of crude mixtures of both brominated products. The following variations can be used.

Processing b.

The simplest method is to adjust the pH of the aqueous solution of the crude product to 2.5 or above while possibly removing the low-boiling organic solvent.

Processing c.

A somewhat more complicated method involves removing the low boiling solvent in vacuo at pH 2.5, extracting the desired products with ethyl acetate at pH 2.5 and washing the combined extracts once with a small volume of saturated aqueous sodium chloride solution and mixing the combined extracts with water.

Processing d.

In this method, the reaction mixtures are usually treated to compare the effects of variations in reaction conditions with the yield of the desired brominated products in the same manner as in treatment a. The finally obtained combined extract is washed with saturated sodium chloride solution, dried over anhydrous inorganic salt, filtered and evaporated in vacuo. Purification extraction with dichloromethane at approximately neutral pH can be omitted when the conditions used are known to give relatively good yields and the excipient used is not extracted at pH 2-4 with, for example, ethyl acetate.

The combined yield of the compounds of formula I and II of the mixed products obtained in treatment d is evaluated by a semi-quantitative method. The purely theoretical maximum yield is calculated using the integrals of the well-separated absorption lines of the PMR spectrum. This determines the ratio of dibromide to monobromide and the detectable known impurities, which are the residual solvent, often ethyl acetate, sometimes the residual excipient and possibly water.

The estimated actual reaction yield is therefore a qualitative estimate based generally on the appearance of the spectrum. As a result, the difference between the maximum yield and the actual yield applies to quality assessments that are excellent, very good, good and reasonable. When taking the PMR spectra of the mixtures of the weighed amount of the separated product and the reference compound, i.e.

Of the weighed amount of 3,4,5-trimethoxyphenylacetic acid, it has been repeatedly found that the actual reaction yield was always at least 90% of the calculated actual yield. On average, from time to time, an overestimation of the actual yield is compensated for by correspondingly ignoring the water content of the starting material of the formula III (6-7% by weight).

I! 19 80038

The process described above for the conversion of β-β-amino-penicillanic acid 1,1-dioxide by diazotization bromination to compounds of formula II and / or formula I using relatively gentle and variable conditions involving the formation of a diazonium intermediate is in principle also relatively easy to apply. amino-PAS-esters. However, esters of compounds of formulas I and II, especially esters of formula II, which can be separated by such reactions, are more conveniently prepared by a process described below comprising novel 6-diazo intermediates as intermediates, as the very gentle conditions of this second invention are more suitable for the fragile nature of the ester groups.

The diazotization-bromination process of the invention described above has hitherto been described in particular in connection with the amino-PAS compound. However, it can also be used in connection with other similar compounds belonging to the general formula III, and in particular in connection with those in which no esters or salts of the formula III are present.

Another embodiment of the present invention comprises, as novel intermediates, 6-diazopenicillanic acid-1,1-dioxide derivatives of the general formula:

“'/ CH_X

: / \ /

N N - C -C

III '' CH

: x C - N-CH J X

: · «S

CO-OR

wherein X is as defined above and R is hydrogen or an ester radical, or when R is hydrogen, its salts.

Such diazo compounds can then be separated and stored because they are sufficiently thermally stable under suitable storage conditions and are then suitable for various treatments, such as bromination to compounds of formula I and / or II.

20 80038

It should be noted that the compounds of general formula X per se and their salts form an important and distinct embodiment of the present invention. As another independent indication, after diazotization of the compounds of general formula III, they behave surprisingly and unexpectedly differently from 6-APA, and the corresponding derivatives, such as 6-diazopenicillanic acid, have not previously been isolated or used crude. in the following reactions.

This aspect of the invention is particularly described below with reference to diazo compounds of formula X wherein X and R are hydrogen atoms.

A compound of formula X (X and R are hydrogen) is prepared simply by dropwise addition of at least an equimolar amount of a commercially available cycloalkyl nitrite or alkynitrite to a suspension of a compound of formula III (wherein R is hydrogen) in a suitable organic solvent to give a relatively strongly acidic carboxyl radical a reaction catalyst necessary to effect a reaction between the nitrite group of the nitrating agent and the amino group of the substrate. Almost quantitative conversion to the 6-diazo derivative of formula X (wherein X and R are hydrogen) can be detected by complete dissolution of the substrate. The reaction proceeds very well regardless of whether or not the small amount of water present in the starting compound and the solvent used has been removed. As a result, it is preferable to work with a non-pre-dried, commercially available, good quality organic solvent and starting material in a relatively stable form containing 6 to 7% by weight of water. The reaction temperature may be as high as about 25 ° C, but it is preferable to use a temperature range between 0 and 10 ° C. Using conventional methods, a compound of formula X (wherein X and R are hydrogen atoms) can be isolated as such, but it is more preferable to obtain this compound as a sodium or potassium salt by adding the corresponding alkali metal salt of a weakly acidic organic compound such as <* -ethyl]. The separated products are in crude form because they usually contain varying but small amounts of solvent and / or cycloalkanol or alkanol derived from organic nitrite and / or, for example, α-ethyl-caproic acid. However, no decomposition products are present. Actual yields can be as high as 70% or higher. Although hydroxyl-containing solvents such as ethanol and reactive carbonyl solvents such as ketones and aldehydes are not suitable for use in this reaction, the other solvents mentioned above, such as acetonitrile, tetrahydrofuran and dioxide, are suitable for achieving a very good reaction, although easy separation of the salt of diazo-penicillanic acid 1,1-dioxide, the treatments are easier to carry out when methyl acetate or ethyl acetate is used as the reaction solvent.

The conditions suitable and practical for converting the esters of compound III to the corresponding 6-diazo derivatives of formula X are the same as described above, except that in order to obtain a rapid reaction it is necessary to add a preferably water-soluble inert organic compound having sufficient acidity and a pKa value of 1.5-5.0 in water, for example preferably a water-soluble carboxylic acid which can be completely removed by one or two washing treatments of the resulting solutions of the compounds of the formula X. A suitable relative amount of the acidic substance which is preferably added before the addition of the (cyclo) alkyl nitrite depends on the nature of this acidic substance, i. its acidity. This relative amount may therefore be between a few mole percent and one acid equivalent relative to the amount of substrate. The organic acid may be as weak as acetic acid, but it is preferable to use a relatively strong carboxylic acid such as chloroacetic acid or oxalic acid. The progress of the reaction is monitored by thin layer chromatography or, if the nature of the solvent is appropriate, by infrared spectroscopy. After the complete reaction, the resulting solutions are washed directly when the solvent used is, for example, ethyl acetate, 22 80038 once or twice with water containing sufficient salt to operate at about -10 ° C, or similarly washed with a water-soluble solvent such as dichloromethane or ethyl acetate. followed by normal further processing. The esters of 6-diazopenicillanic acid 1,1-dioxide finally obtained, of formula X, are usually solids, sometimes crystalline substances. The yield of crude products is from good to very good, and the purity of such preliminary products varies from moderate to good.

The diazo compounds of formula X thus obtained can be used to convert them into compounds of formula I and / or II. In the case where R is hydrogen, bromination usually provides a mixture containing both brominated compounds of formulas I and II. However, in the case of esters, bromination usually provides monobromides of formula I (where R is an ester residue) almost completely without esters of dibromides of formula II, or dibromides of formula II with almost no corresponding monobromides of formula I, depending on whether or not the carboxylic acid was removed. used in the preparation of intermediates of formula X by washing with water or left in a solution containing intermediates of formula X, which shows the unique features of this invention, since esters of 6-diazopenicillanic acid 1,1-dioxide can be modified by this preparation method 6a to pure esters of bromo-penicillanic acid 1,1-dioxide or 6,6-dibromo-penicillanic acid 1,1-dioxide.

Although separated diazo compounds of formula X may be used to prepare compounds of formula I and / or II, it is, of course, usually more preferred to use in situ preparation of diazo derivatives. When R is hydrogen, there is one treatment. When R is an ester radical, there are two possible treatments.

When R is hydrogen, bromine and hydrobromic acid are usually added after 6-diazopenicillanic acid 1,1-dioxide has been added.

II

23 8 0 0 3 8 formed in situ. In these preparations, which are carried out in situ, the organic solvent is preferably acetonitrile, 1,1-dimethoxyethane or, for example, ethyl acetate together with up to 20% by volume of sulfolane, dimethyl sulfoxide, tetramethylurea or a non-hydroxylated derivative of malonic acid as defined above.

The following non-limiting examples and methods of preparation illustrate the present invention.

Example I

Preparation of a crude mixture of 6,6-dibromo-penicillanic acid 1,1-dioxide and 6α-bromo-penicillanic acid 1,1-dioxide 2.9 ml of concentrated sulfuric acid were carefully added to 25 ml of ice water. 3 ml of the dilute sulfuric acid thus prepared was added to 30 ml of acetonitrile at 0-5 ° C. The amount of sulfuric acid used is therefore about 10.5 mmol, or 21 acid equivalents. With continuous operation at a temperature between 0-5 ° C, 2.0 g (assumed to be equivalent to 8.0 mmol; in fact only about 7.5 mmol) of 68-aminopenicillanic acid 1,1-dioxide were added with stirring. This did not result in complete dissolution. 8 g (25 mmol) of pyridine hydrobromide perbromide (CH 2 Cl 2) were then added to give a clear solution immediately. 660 mg of sodium nitrite (9.5 mmol) were added in five, approximately equal doses over about 15 minutes and then stirred for another 30 minutes.

A solution of 1 g of sodium metabisulfite in 20 ml of water was then added with stirring, after which the pH was raised to 5.0 by the careful addition of 4N sodium hydroxide. The acetonitrile was removed azeotropically by concentration: at a pressure of about 15 mm Hg for about 10 min. The resulting solution was acidified with 4N

hydrochloric acid to pH 2.4 and then extracted three times with approximately equal volumes of ethyl acetate. The combined extracts were dried over anhydrous magnesium sulfate, filtered and completely evaporated in vacuo. After thorough drying in vacuo, the resulting solid weighed 2.61 g.

24 80038

According to the PMR spectrum of the product (d - DMSO 60 Mc) it contained

O

6,6-dibromopenicillanic acid 1,1-dioxide, 6a-bromo-penicillanic acid 1,1-dioxide, ethyl acetate and pyridine (pyridine itself and / or pyridine.HCl or pyridine.HBr) in a molar ratio of 13: 1: 1, 5: 3.5. Assuming the presence of pyridine as such, the calculated maximum yield was 78.7% of usable products. Given the very small amount of degradation products present as determined by PMR spectrum and thin layer chromatography, and the possibility that pyridine is present as a salt, the actual yield was calculated to be about 65-70%.

b. The experiment was repeated in the same way, but with one change in the separation treatment. To remove the pyridine-containing substances, the ethyl acetate extract containing both bromides was washed twice with a small volume of saturated sodium chloride solution. The resulting mixture of bromides weighed 2.46 g, giving a molar ratio of dibromide to monobromide of 8: 1. Given the small amount of ethyl acetate remaining, the maximum yield was calculated to be 78.2% of usable products. Considering good to very good quality in terms of Degradation Products, 011 actual yield 70% or better.

c. This experiment was repeated, but using one variation. Instead of 25 mmol, only 12 mmol (3.84 g) of brominated complex was used, which generally constitutes a more suitable excess. The yield of the crude bromide mixture was now 2.39 g of significantly less pure product containing, in addition to a small amount of ethyl acetate, dibromide, monobromide and an unknown by-product in a molar ratio of 75:13:12. Apparently, the lower amount of hydrobromic acid, because it was substantially compensated for by pyridine, was not sufficient in this experiment to compensate for the reduced acidity of the second acid equivalent of sulfuric acid.

Example II

Conversion of 68-amino-penicillanic acid 1,1-dioxide to a mixture of 6,6-dibromo and 6α-bromo-penicillanic acid 1,1-dioxide_

Continuously operating at a temperature between 0 and 8 ° C maintained by external cooling of crushed ice

II

Using 80 g of water, 2.0 g of 6β-amino-penicillanic acid 1,1-dioxide was added to a stirred solution of 2.0 g (10.5 mmol) of p-tolylsulfonic acid monohydrate in a mixture of 3 ml of water and 30 ml of acetonitrile to give complete dissolution. After dissolving 8.0 g (25 mmol) of pyridine hydrobromide perbromide, 660 mg (9.5 mmol) of sodium nitrite were added as described in Example I. The reaction mixture was treated in the same manner, washing the extract with concentrated sodium chloride solution. Yield 2.54 g.

Thin layer chromatography (TLC) and PMR spectra showed good or very good purity of the separated product with respect to two unknown degradation products present in small or very small amounts. Excluding decomposition products, the PMR spectrum showed that the molar ratio of dibromide, mono-bromide, ethyl acetate to acetic acid was 12.5: 1: 1: 1.4. The calculated maximum yield was 79.8%. The actual yield was calculated to be about 72%. This was found after reduction with zinc at a temperature of 0-5 ° C and a pH ranging from 4.2 to 4.7 to give penicillanic acid 1,1-dioxide in at least 60% yield, which was present in a truly pure final product. in terms of.

Example III

Repeat of Example II, again containing p-tolylsulfonic acid, but using some modifications_ (a) The experiment of Example II was repeated on the same scale and under the same reaction conditions, but the acetonitrile was now replaced by the same volume of ethyl acetate. Since the yield and purity of the crude bromide mixture could be expected to be less good, the aqueous solution was extracted after treatment with bisulfite and after removal of ethyl acetate in vacuo initially at pH 6-6.5 twice with dichloromethane. The extraction of the desired products at pH 2.0 with ethyl acetate was also adjusted to some extent by TLC. The yield of the product was 2.17 g and its quality was good by TLC and PMR analyzes.

The ratio of 6,6-dibromo-penicillanic acid 1,1-dioxide to 6α-bromo-penicillanic acid 1,1-dioxide was now 9: 5, indicating that this ratio also depends on the nature of the organic solvent used. Taking into account the amount of ethyl acetate present, the maximum yield was calculated to be 61.3%. Given the relatively higher amount of unknown degradation products, the actual yield was estimated to be 50-55%.

(b) The fact that bromide anions must be present in order to obtain a satisfactory yield was found in three experiments in which hydrobromic acid was omitted but higher amounts of p-tolylsulfonic acid were used. In the first experiment, 2.0 g of 68-aminopenicillanic acid 1,1-dioxide, 4.0 g (21 mmol) of p-tolylsulfonic acid, 0.95 ml of pyridine (12 mmol), 0.62 ml (12 mmol) of ) bromine, 660 mg (0.5 mmol) of sodium nitrite, 30 ml of acetonitrile and 3 ml of water. The yield was only 1.03 g. Monobromide formed the main product and the estimated actual yield was only about 24%.

In the other two experiments, the ingredients were 1.82 g (7.35 mmol) of 68-aminopenicillanic acid 1,1-dioxide, 4.18 g (22 mmol) of p-tolylsulfonic acid, 12 mmol of pyridine and bromine, 9.5 g. mmol of sodium nitrite, 25 ml of acetonitrile and 2.7 ml of water. With yields of 1.20 g and 1.02 g, the results were also poor. In both cases, the main product was monobromide and the estimated actual yields were less than 20%.

Example IV

Repetition of Example I (a) under changed conditions (a) In the same manner as in the above examples, 2.0 g (8 mmol) of 68-aminopenicillanic acid 1,1-dioxide were treated in a mixture of 30 ml of acetonitrile, 3 ml of water and 5 ml of , 25 mmol of sulfuric acid (10.5 acid equivalents), 3.84 g (12 mmol) of pyridine hydrobromide perbromide and 660 mg (9.5 mmol) of sodium nitrite. The aqueous layer obtained after the usual procedures was first purified to some extent by extraction with dichloromethane at pH 6.0-6.5 and then extracted at pH 2.0 with ethyl acetate. The product yield was 2.06 g. In view of the presence of ethyl acetate and 6,6-dibromo and 6a-bromo-penicillanic acid 1,1-di-

II

27 The 80038 oxide molar ratio of 9:11 was calculated to have a maximum yield of 69.6%. Considering the relatively good quality with respect to the unknown Decomposition Products, the actual yield was estimated to be 59-61%.

(b) Experiment (a) was repeated using otherwise identical conditions and amounts of material but ethyl acetate as the organic solvent. The yield was 1.87 g and the dibromide / monobromide molar ratio was 9: 5. The calculated maximum yield was 61.3%. As the quality of the product was only satisfactory, the actual yield was presumably not better than about 50%.

Example V

Conversion of 68-amino-penicillanic acid 1,1-dioxide to a mixture of 6,6-dibromo and 6α-bromo-penicillanic acid 1,1-dioxide (a) At 0-5 ° C, 1.2 ml of 47% was added: hydrobromic acid (10.5 mmol) and 1.3 ml of water in 25 ml of acetonitrile. Addition of 2.0 g (8 mmol) of 68-aminopenzylic acid 1,1-dioxide gave a clear solution. Working continuously at a temperature below 5 ° C, 3.84 g (12 mmol) of pyridine hydrobromide di-bromide was added, followed immediately by 660 mg (9.5 mmol) of sodium nitrite in approximately five equal portions over about 15 minutes. After stirring for a further 30 minutes, 1 g of sodium metabisulphite in 20 ml of water was added, after which the pH was raised to about 5.0 by the careful addition of 4N sodium hydroxide. The acetonitrile was removed by concentration in vacuo. The aqueous solution was extracted at pH 2 with ethyl acetate, the combined extracts were washed once with a small volume of saturated sodium chloride solution, dried over anhydrous magnesium sulfate, filtered and completely evaporated in vacuo. After drying in vacuo, the yield was 2.27 g of product, the quality of which was good to very good by TLC and PMR analyzes in terms of the presence of unknown decomposition products.

The PMR spectrum showed that the molar ratio of dibromide, monobromide to ethyl acetate was 8: 6.5: 2.3. The isolated product 28 80038 therefore contained 1.325 g (3.388 moles) of 6,6-dibromo-penicillanic acid 1,1-dioxide and 0.859 g (2.753 mmol) of 6α-bromopenicillanic acid 1,1-dioxide, and thus a total of 6/141 mmol of useful product. As a result, the maximum yield was 76.7% and the actual yield was estimated to be at least 70%.

(b) Experiment (a) was repeated using two transformations. While maintaining the ratio of acetonitrile to water at 10: 1, the amount of hydrobromic acid was added to 14.5 mmol and the aqueous solution was first purified to some extent by extraction with dichloromethane at pH 6.5 before extraction of the useful product at pH 2.0. The product of very good quality weighed 2.30 g. The ratio of dibromide to monobromide was 5.8: 4. The maximum yield was 77.8% and the estimated actual yield was at least 72%.

(c) Experiment (a) was repeated using 7.68 g (24 mmol) of pyridine hydrobromide perbromide with a separation treatment identical to that shown in Experiment (b). The separated product, which had good to very good quality, weighed 2.30 g. The ratio of dibromide to monobromide was 9.7: 6. The maximum yield was 76.8% and the estimated actual yield was 68-72%.

Example VI

Conversion of 68-amino-penicillanic acid 1,1-dioxide to a mixture of 6,6-dibromo and 6α-bromo-penicillanic acid 1,1-dioxide_ (a) Working continuously at 0-5 ° C, 0.62 ml (12 mmol) was added. ) bromine to a solution of 1.82 g (7.35 mmol) of 68-aminopenicillanic acid 1,1-dioxide in a mixture of 22.5 ml of acetonitrile, 2.5 ml of dimethyl sulfoxide, 2.5 ml of water and 22 mmol of hydrobromic acid (water and hydrobromic acid were added as 2.5 ml of a 47% aqueous acid solution), followed by the portionwise addition of 660 mg of sodium nitrite as described above. The conventional method for separating the product involved purifying the aqueous solution with dichloromethane at pH 6.5. Yield 2.08 g of product of at least good quality according to the PMR spectrum in terms of unknown decomposition products

II

29 80038 impurities. This product contained 6,6-dibromo-penicillanic acid 1,1-dioxide, 6α-bromo-penicillanic acid 1/1 dioxide, ethyl acetate, dimethyl sulfoxide and water in a molar ratio of 12.3: 1.2: 1.3: 0.5: 4. The maximum yield was calculated to be 70.6% and the calculated actual yield to be 61-63%.

The experiment was repeated using 22.5 ml of ethyl acetate instead of the same volume of acetonitrile. The yield was 1.58 g. The calculated maximum yield was 53.9% and the molar ratio of dibromide, monobromide, ethyl acetate, dimethyl sulfoxide to water was 12: 1.5: 1.5: 0.5: 3. Considering the excellent quality of the product with respect to the unknown Decomposition Products, the actual yield was estimated to be 50%.

(b) Working continuously at 0-5 ° C, 0.62 ml of bromine was added to a solution of 1.91 g (7.7 mmol) of 6β-aminopenicillanic acid 1,1-dioxide in a mixture of 22.5 ml. ethyl acetate, 2.5 ml of tetrahydrothiophene-1,1-dioxide (sulfolane) and 2.5 ml of a 47% aqueous solution of hydrogen bromide (corresponding to 22 mmol of hydrobromic acid), followed by the portionwise addition of 660 mg (9.5 mmol) of sodium nitrite as described above. way. The conventional method for separating the product involved purifying the aqueous solution with dichloromethane at pH 6.5. Yield 2.17 g of product which, on the basis of the PMR spectrum, was of relatively good quality with regard to the unknown decomposition products present. The calculated maximum yield of useful product was 60.7% and the molar ratio of dibromide, monobromide, sulfolane to water was 10: 0.6: 6: 6. The actual yield was estimated to be 53-55%.

(c) This experiment was performed in the same manner starting from 2.0 g (8 mmol) of 66-aminopenicillanic acid 1,1-dioxide, 0.62 ml (12 mmol) of bromine, 660 mg (9.5 mmol) of bromine. ) sodium nitrite, 22 mmol hydrogen bromide and 2.5 ml water (added as 2.5 ml 47% aqueous hydrogen bromine), and a mixture of 22.5 ml acetonitrile and 2.5 ml 30 80038 was used as the organic solvent. (23.5 mmol) ethyl cyanoacetate. The separation treatment was similar to (b). Yield 2.40 g of product of good quality based on the PMR spectrum with respect to the unknown Decomposition Products present. The calculated maximum yield was 77.0% and the molar ratio of dibromide, monobromide, ethyl acetate to water was 6: 5.5: 3.3: 7. The actual yield was estimated to be 69-71%.

(d) In the same manner, an experiment was performed using 8 mmol of 68-aminopenicillanic acid 1,1-dioxide, 12 mmol of bromine, 2.5 ml of 47% aqueous hydrogen bromine solution (22 mmol of HBr), 9.5 mmol of sodium nitrite, 22 mmol of , 5 ml of ethyl acetate and 2.5 ml of ethyl cyanoacetate. The product yield was 2.16 g.

The quality was at least good. Dibromide: monobromide: ethyl acetate ratio = 10.9: 4.6: 5.3. Maximum yield 67.9%. Calculated actual yield 60-62%.

(e) In a similar manner, an experiment was performed using 8 mmol of 6g-aminopenicillanic acid 1,1-dioxide, 12 mmol of bromine, 2.5 ml of 47% aqueous hydrogen bromine solution (22 mmol of HBr), 9.5 mmol of sodium nitrite, 22, 5 ml of acetonitrile and 2.5 ml of diethyl malonate. The product yield was 2.23 g. The quality was very good. Dibromide: monobromide: ethyl acetate: diethyl malonate ratio = 6.2: 6.5: 4: 1.7. Maximum yield 69.7%. Estimated actual yield 64-67%.

Example VII

Conversion of? -amino-penicillanic acid 1,1-dioxide to a mixture of 6,6-dibromo and 6α-bromo-penicillanic acid 1,1-dioxide_ (a) The reaction was carried out repeatedly as described above using 2.0 g (8 mmol) 68 -amino-penicillanic acid 1,1-dioxide, 22 mmol hydrobromic acid, only 200 mg N-benzoyl-benzenesulfonamide (about 0.77 mmol), 0.62 ml (12 mmol) bromine, 660 mg sodium nitrite (9.5 mmol), 25 ml of acetonitrile and 2.5 ml of water. The product was separated by a multi-step treatment which included washing the aqueous layer with dichloromethane at pH 6.5. Yield 2.44 g of product having a minimum of PMR spectrum

II

31 80038 good quality for unknown Decomposition products. The calculated maximum yield was 72% and the molar ratio of 6,6-dibromo-penicillanic acid 1,1-dioxide, 6α-bromo-penicillanic acid 1,1-dioxide, ethyl acetate and N-benzoylbenzenesulfonamide was 10.6: 2.7: 2: 1.8. The actual yield was estimated to be 64-67%.

(b) Experiment (a) was repeated, but instead of the sulfonamide, 0.2 ml (about 1.88 mmol) of ethyl cyanoacetate was used as an excipient. Yield 2.45 g of product of at least good quality based on the PMR spectrum in terms of the presence of unknown decomposition products. The calculated maximum yield was 75.8%, taking into account the molar ratio of dibromide, monobromide and ethyl acetate of 13.6: 2.2: 2.3, and the actual yield was estimated to be 68-70%.

Example VIII

Effect of the nature of the excipient on the yield of useful products for diazotization-bromination of the ratio of 68-amino-penicillanic acid 1,1-dioxide and 6,6-dibromo-penicillanic acid 1,1-dioxide to 6a-bromopenicillanic acid 1,1-dioxide in this reaction_

The results of the excipient variation are presented in the following abbreviated form. All reaction mixtures were treated in the same manner. After completion of the reaction, dilute sodium bisulfite was added, followed by partial neutralization by the addition of 4N sodium hydroxide, removal of the low-boiling organic solvent, usually acetonitrile, and purification extraction with dichloromethane at pH 6.0-6.5, depending on the number of these extractions. the nature of the excipient used, three extractions of the useful products with ethyl acetate at pH 2.0, two washing of the combined acidic extracts with a small volume of saturated aqueous sodium chloride solution, drying the obtained extract over anhydrous magnesium sulphate, filtration, evaporation in vacuo and finally drying. The separated products were analyzed by proton magnetic resonance (PMR) spectrum using solutions in dg-DMSO at 60 Mc and tetramethylsilane as internal reference. All reactions 32 80038 were performed at 0-5 ° C. Nitration was always performed using sodium nitrite, which was added in five approximately equal portions over about 15 minutes and then stirred for an additional 30 minutes.

These abbreviated examples show the amount of 68-amino-penicillanic acid 1,1-dioxide (denoted by the abbreviation "amino-PAS") excluding the amount of water present, all ingredients in mmol or volume, yield in grams, molar ratios between known components in the separated product, the calculated maximum yield, excluding the presence of known degradation products, the quality of the separated product with respect to these unknown degradation products and the estimated actual yield of usable products.

As mentioned above, the actual yield has been calculated on the basis of the maximum yield and the quality of the product, the latter being determined by means of a PMR spectrum and thin-layer chromatograms.

(a) Reaction in the presence of 2,4,6-trimethylpyridine (collidine) 8 mmol amino-PAS, 22 mmol HBr, 12 mmol collidine, 12 mmol bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and 2.5 ml water. Yield 1.80 g. Quality: good. Dibromide: mono-bromide: ethyl acetate = 10.5: 4.0: 2.7. Maximum yield 58.4%. Estimated actual yield about 50%.

(b) Using 1-methylimidazole 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 1-methylimidazole, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile, 2.5 mL water. Yield 2.07 g. Quality: excellent. Dibromo-di: monobromide: ethyl acetate = 11: 7: 1.3. Maximum yield 70.5%. Estimated actual yield of at least 67%.

(c) Using 1-methylimidazole

Experiment (b) was repeated using one variation: instead of 12 mmol, only 9 mmol of bromine was now used. Yield 2.21 g.

Il 33 80038

Quality: reasonable. Dibromide: monobromide: ethyl acetate = 1: 8.5: 10.5. Maximum yield 66.2%. Estimated actual yield about 55%.

(d) Using 1-methylimidazole

Experiment (b) was repeated using 1,2-dimethoxyethane (25 mL) as the organic solvent and more NaNO 2 (11 mmol instead of 9 mmol). Yield 2.21 g. Quality: relatively good. Dibromide: monobromide: ethyl acetate = 6.5: 3.6: 1.2. Maximum yield 73.1%. Estimated actual yield 60-63%.

1) An experiment carried out on the same scale as in (d) but comprising a normal amount of NaNO 2 (i.e. 9.5 mmol) and only 9 mmol of bromine gave a less good result than experiment (c): 1.83 g of product of only reasonable quality, maximum yield 65.7% almost alone monobromide, the actual yield was estimated to be about 50-52%.

2) An experiment similar to (d), but using only 10.5 mmol of HBr, gave 1.84 g of product of only reasonable quality, a ratio of dibromide to monobromide of 1:10, and an estimated actual yield of less than 45%. In these experiments, the acidity is apparently too low, especially since 1-methylimidazole is not a weak base.

(e) Using 3-methylpyridine, 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 3-methylpyridine, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile and 2.5 mL water. Yield 1.98 g. Quality: good. Dibromide: monobromide: ethyl acetate = 5: 8: 2. Maximum yield 69.5%. Estimated actual yield 60-62%.

(f) Using 2-dimethylaminopyridine 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 2-dimethylaminopyridine, only 9 mmol bromine, 9.5 mmol NaNO 2, 34,80038 25 ml acetonitrile and 2.5 ml water. Yield 1.81 g. Quality: excellent. Monobromide: ethyl acetate = 10.5: 15. Maximum yield 51.8%. Estimated actual yield 50% or greater.

(g) Using 4-dimethylaminopyridine, 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 4-dimethylaminopyridine, only 9 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile and 2.5 mL water. Yield 1.87 g. Quality: very good. Dibromide: monobromide: ethyl acetate = 0.7: 11.0: 15.5. Maximum yield 53.9%. Estimated actual yield of at least 50%.

(h) Using 2,6-dimethylpyridine 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 2,6-dimethylpyridine, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile, 2.5 mL water . Yield 1.92 g. Quality: good. Dibromide-monobromide: ethyl acetate = 11: 4: 1.7. Maximum yield 63.1%. Estimated actual yield about 55%.

(i) Using 3-methylisoquinoline 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 3-methylisoquinoline, only 9 mmol bromine, 9.5 mmol NaNC> 2, 25 mL acetonitrile and 2.5 mL water. Yield 1.73 g. Quality: from good to very good. Maximum yield 63.9%. Estimated actual yield 55-57%. Dibromide-monobromide: ethyl acetate = 2.3: 9.8: 1.6.

(j) Using 4-methylpyridine-1-oxide 8 mmol amino-PAS, only 10.5 mmol HBr, 12 mmol 4-methylpyridine 1-oxide, 12 mmol bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and only 1.3 ml of water. Yield 1.83 g. Quality: relatively good. Dibromide: monobromide: ethyl acetate = 8: 5: 1.2. Maximum yield 62.1%. Estimated actual yield 50% or more.

(k) Using triethylamine 8.0 mmol amino-PAS, 22 mmol HBr, 12 mmol triethylamine, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile

II

35,80038 reels and 2.5 ml water. Yield 1.72 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: water = 11.6: 1.8: 1.7. Maximum yield 53.6%. Estimated actual yield about 48%.

(l) Using 5,5-dimethylhydantoin

Experiment (1) —1: 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 5,5-dimethylhydantoin, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile, and 2.5 mL water. Yield 2.26 g. Quality: good. Dibromide: monobromide: ethyl acetate: dimethylhydrantoin: water = 3.1: 2: 1.8: 1: 4. Maximum yield 67.4%. Estimated actual yield 58-60%.

Experiment (1) -2: 8 mmol amino-PAS, only 11 mmol HBr, 12 mmol 5,5-dimethylhydantoin, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile, and 2.5 mL water. Yield 2.30 g. Quality: relatively good. Dibromide: monobromide: ethyl acetate: dimethylhydantoin: water = 4: 3: 1.9: 1.2: 5. Maximum yield 69.2%. Estimated actual yield about 60%.

(m) Using 2-methylpyrazine 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 2-methylpyrazine, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile and 2.5 mL water. Yield 1.56 g. Quality: good. Mono-bromide: ethyl acetate: 2-methylpyrazine: water = 8.8: 1.8: 1.8: 2. Maximum yield 54.8%. Estimated actual yield about 50%.

A very similar experiment using only 11 mmol HBr yielded, in approximately the same yield and purity, a product containing, as expected, a molar ratio of dibromide and mono-bromide in a molar ratio of 2.3: 4.

(n) Using N-phenyl-phenylcarbonamide

Experiment (n) -1: 8 mmol amino-PAS, 22 mmol HBr, 12 mmol N-phenylphenylcarbonamide, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile, 2.5 mL water. Yield 2.56 g.

36 80038

Quality: from good to very good. Dibromide: monobromide: ethyl acetate: N-phenyl-phenylcarbonamide = 10: 1.5: 1.4: 4: 3. Max mismatch 68.8%. Estimated actual yield 62-64%.

Experiment (n) -2: Similar to Experiment (n) -1 except that only 11 mmol HBr was used. Yield 2.40 g.

Quality: relatively good. Dibromide: monobromodiethyl acetate: N-phenyl-phenylcarbonamide = 5: 5.5: 3.0: 3.7. Maximum yield 67.5%. Estimated actual yield about 60%.

(o) Using N-ethyl-piperidine 8 mmol amino-PAS, 22 mmol HBr, 12 mmol N-ethylpiperidine, 12 mmol bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and 2.5 ml water. Yield 1.61 g. Quality: very good. Dibromide: monobromide: ethyl acetate = 12: 2: 1.8. Maximum yield 51.5%. Estimated actual yield about 48%.

(p) Using phthalimide

Experiment (p) -1: 8 mmol amino-PAS, only 11 mmol HBr, 12 mmol phthalimide, 12 mmol bromine, 9.5 mmol NaNCl 2, 25 ml acetonitrile and 2.5 ml water. Yield 2.04 g. Quality: relatively good. Dibromide: monobromide: ethyl acetate: phthalimide-water = 7: 3.5: 1.7: 0.5: 3. Maximum yield 65.0%. Estimated actual yield about 55%.

Experiment (p) -2: Identical to Experiment (p) -1 except that 22 mmol of HBr was now used. Yield 2.0 g. Quality: very good. Dibromide: monobromide: ethyl acetate: phthalimide = 11.2: 2.2: 0.7: 0.8. Maximum yield 64.4%. Estimated actual yield 60%.

(q) Using N-methylmorpholine

Experiment (q) —1: 8 mmol amino-PAS, 22 mmol HBr, 12 mmol N-methylmorpholine, 12 mmol bromine, 9.5 mmol NaNO 2 # 25 mL acetonitrile, and 2.5 mL water. Yield 2.0 g. Quality: good. Dibromide: monobromide: ethyl acetate = 10: 4: 1.3. Maximum yield 68.9%. Estimated actual yield 61-63%.

il 37 80038

Experiment (q) -2: In contrast to Experiment (q) -1, only 9 nmoles of bromine and 1,2-dimethoxyethane were used instead of acetonitrile. Yield 2.01 g. Quality: relatively good. Dibromide: monobromide: ethyl acetate: dimethoxyethane = 5: 6.4: 2.3: 0.25. Maximum yield 68.6%. Estimated actual yield 58-60%.

(r) Using hydantoin, 8 mmol amino-PAS, 22 mmol HBr, 12 mmol hydantoin, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile, and 2.5 mL water. Yield 2.18 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: hydantoin = 11.5: 1: 1.3: 0.5. Maximum yield 68.4%. Estimated actual yield 62-64%.

(s) Using phenylacetylurea 8 mmol amino-PAS, 22 mmol HBr, 12 mmol phenylacetylurea, 12 mmol bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and 2.5 ml water. Yield 2.07 g. Quality: good. Dibromide: monobromide: ethyl acetate: phenylacetylurea: water = 14.2: 1.3: 0.7: 0.6: 5. Maximum yield 64.5%. Estimated actual yield 56-58%.

(t) Using phenylacetamide

Experiment (s) -1: 8 mmol amino-PAS, 22 mmol HBr, 12 mmol phenylacetamide, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile and 2.5 mL water. Yield 2.25 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: phenylacetamide = 10: 1.2: 1.2: 1.0. Maximum yield 69.3%. Estimated actual yield 62-64%.

Experiment (s) -2: In this experiment, somewhat less starting material and somewhat more water were used, while no attempt was made to remove the excipient more or less quantitatively. 7.35 mmol amino-PAS, 22 mmol HBr, 12 mmol phenylacetamide, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile and 2.7 mL water. Yield 2.42 g. Quality: from good 38 to 80038 to very good. Dibromide monobromide: ethyl acetate: phenylacetamide: water = 7.5: 2: 1: 5.5: 4. Maximum yield 70.2%. Estimated actual yield 63-65%.

(u) Using acetamide 8 mmol amino-PAS, 22 mmol HBr, 12 mmol acetamide, 12 mmol bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and 2.5 ml water. Yield 1.57 g. Quality: very good. Dibromide: monobromide: ethyl acetate: acetamide: water = 9.5: 3: 1.5: 0.4: 9. Maximum yield 49.4%. Estimated actual yield 45% or more.

(v) Using N-phenylacetamide 7.35 mmol amino-PAS, 22 mmol HBr, 12 mmol N-phenylacetamide, 12 mmol bromine, 9.5 mmol NaNO 2 # 25 mL acetonitrile and 2.7 mL water. Yield 1.6 g. Quality: very good. Dibromide: monobromide: ethyl acetate: N-phenylacetamide: water = 12: 1.4: 0.9: 0.8: 4. Maximum yield 56.2%. Estimated actual yield 50-52%.

(w) Using pyridyl-4-carbonamide 7.35 mmol amino-PAS, 10.5 mmol HBr, 10 mmol p-tolylsulfonic acid, 12 mmol pyridyl-4-carbonamide, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile and 2.7 ml of water. Yield 2.00 g. Quality: relatively good. Dibromide: monobromide: ethyl acetate: water = 10: 1.6: 1: 8. Maximum yield 68.7%. Estimated actual yield 54-56%.

(x) Using N-benzoylbenzenesulfonamide 8 mmol amino-PAS, 22 mmol HBr, 12 mmol N-benzoylbenzenesulfonamide, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile and 2.5 mL water. Yield 2.67 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: N-benzoyl-benzenesulfonamide = 4.3: 0.7: 0.5: 0.5. Maximum yield 78.6%. Estimated actual yield 71-73%.

(y) Using p-tolylsulfonamide 8 mmol amino-PAS, 22 mmol HBr, 12 mmol p-tolyl

II

39 80038 sulfonamide, 12 mmol bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and 2.5 ml water. Yield 2.35 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: p-tolylsulfonamide: water = 8: 0.5: 1: 2: 4. Maximum yield 63.3%. Estimated actual yield 56-58%.

(z) Using bis (p-tolylsulfone) imide 8 mmol amino-PAS, 22 mmol HBr, 12 mmol bis (p-tolylsulfone) imide, 12 mmol bromine, 9.5 mmol NaNC> 2, 25 mL acetonitrile and 2.5 ml of water. Yield 2.16 g. Quality: good. Dibromide: monobromide: ethyl acetate: bis (p-tolylsulfone) imide = 8.0: 2.0: 0.8: 0.5. Maximum yield 67.7%. Estimated actual yield about 60%.

Example IX

Further examples of the effect of the nature of the excipient on the yield of useful products in the diazotization-bromination treatment of 66-amino-penicillanic acid 1,1-dioxide (see Example VIII) (a) Reaction in the presence of saccharin 8 mmol amino-PAS, 22 mmol HBr, 12 mmol saccharin , 12 mmol of bromine, 9.5 mmol of NaNO 2, 25 ml of acetonitrile and 2.5 ml of water. Yield 2.12 g. Quality: good. Dibromide: mono-bromide: ethyl acetate: saccharin = 8.0: 2.0: 1.0: 1.0. Maximum yield 64.8%. Estimated actual yield 55-57%.

(b) Using 1-methylimidazole 7.7 mmol amino-PAS, 22 mmol HBr, 12 mmol 1-methylimidazole, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL nitromethane and 2.6 mL water . Yield 1.84 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate = 11: 1.4: 1.7. Maximum yield 60.6%. Estimated actual yield 54-56%.

(c) Using succinimide

Experiment (c) -1: 8 mmol amino-PAS, 22 mmol HBr, 12 mmol sea resin imide, only 9 mmol bromine, 9.5 mmol NaNC> 2, 25 mL acetonitrile and 2.5 mL water. Yield 2.21 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate = 3: 13.5: 23. Maximum yield 61.5%. Estimated actual yield 55-57%.

«° 80038

Experiment (c) -2: Similar to Experiment (c) -1 except that 12 mmol of bromine was used. Yield 2.32 g. The quality is good because the product contained only one visible degradation product (about 5-6 mmol%). Dibromide: monobromide: ethyl acetate = 3: 6.5: 3.6. Maximum yield 78.3%. Estimated actual yield 70-72%.

Experiment (c) -3: Identical to Experiment (c) -2 except that 7.35 mmol of amino-PAS and 2.7 ml of water were used. Yield 2.21 g. Quality: very good. Dibromide: monobromide: ethyl acetate: succinimide = 14: 1: 3: 2. Maximum yield 72.1%. Estimated actual yield 67-69%.

Experiment (c) -4: 7.7 mmol amino-PAS, 22 mmol HBr, 12 mmol succinimide, 12 mmol bromine, 9.5 mmol NaNO 2, 22.5 mL ethyl acetate, 2.5 mL sulfolane, and 2.6 mL of water. Yield 1.93 g. Quality: very good. Dibromide: ethyl acetate: sulfolane: succinimide = 10.5: 1.4: 1.7: 2.0. Maximum yield 57.1%. Estimated actual yield 52-53%.

Experiment (c) -5: 7.35 mmol amino-PAS, no HBr, 22 mmol p-tolylsulfonic acid, 12 mmol bromine, 9.5 mmol NaNO 3, 30 mL acetonitrile, 2.7 mL water. The yield was only 1.38 g, if a useful product was present, it was mainly monobromide. Yield estimation did not make sense due to too poor quality.

Experiment (c) -6: Similar to Experiment (c) -2, but the amounts of succinic acid and bromine were increased to 15 mmol and the amount of NaNO 2 to 12 mmol. Yield 2.48 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: succinimide = 16: 0.5: 1.1: 3. Maximum yield 73.0%. Estimated actual yield 66-68%.

Experiment (c) -7: As in Experiment (c) -6, but the amount of succinimide was again brought to 12 mmol. Yield 2.2 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: succinimide: water = 12.2: 0.7: 0.7: 3.5: 6. Maximum yield 66.7%.

il 41 80038

Estimated actual yield 60-62%.

Experiment (c) -8: 8 mmol amino-PAS, 22 mmol HBr, 12 nv mol N-bromosuccinimide, 9/5 mmol NaNO 2/25 ml acetonitrile and 2.5 ml water. Yield 2.09 g. Quality: relatively good. Dibromodirmonobromide: ethyl acetate = 10.5: 4.1: 1.5. Maximum yield 69.1%. Estimated actual yield 59-62%.

Note: It seems reasonable to assume that in the presence of an excess of hydrobromic acid, the N-bromo-succinic acid amide decomposes rapidly to predominantly bromine and succinic acid amide. Thus, the mixture already contains succinic acid imide and bromine in a molar ratio of approximately 10:12:12 HBr before the introduction of NaNO2. An excess of dibromide over monobromide in the separated product confirms such an assumption. Thus, a significantly lower yield of useful products compared to experiment (c) -2, caused by too low an amount of HBr, is also logical.

(d) Using 1-vinylimidazole 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 1-vinylimidazole, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile and 2.5 mL water. Yield 2.44 g. Quality: at least good. Dibromide: monobromide: ethyl acetate = 16.2: 1.6: 1.5. Maximum yield 77.8%. Estimated actual yield of at least 70%.

(e) Using diphenyl sulfoxide 8 mmol amino-PAS, 22 mmol HBr, 12 mmol diphenyl sulfoxide, 12 mmol bromine, 9.5 mmol NaNO 2 »25 mL acetonitrile and 2.5 mL water. Yield 2.00 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: water = 15.2: 1.1: 1.8: 3. Maximum yield 65.2%. Estimated actual yield 59-62%.

(f) Using caprolactam 8 mmol amino-PAS, 22 mmol HBr, 12 mmol caprolactam, 42 80038 12 mmol bromine, 9.5 mmol NaNCl 2, 25 ml acetonitrile and 2.5 ml water. Yield 2.40 g. Quality: at least good. Dibromide: monobromide: ethyl acetate = 11.7: 1.6: 1.6. Maxi yield 78.0%. Estimated yield 69-72%.

(g) Using 1,1,3,3-tetramethylguanidine 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 1,1,3,3-tetramethylguanidine, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mmol acetonitrile and 2.5 ml of water. Yield 2.24 g.

Quality: good. Dibromide: monobromide: ethyl acetate: water = 10.2: 1.6: 1.5: 3. Maximum yield 70.7%. Estimated actual yield 62-64%.

(h) Using pyridine in the reverse method, 8 mmol of amino-PAS was added portionwise at 0-5 ° C over 30 minutes to a mixture of 22 mmol of HBr, 12 mmol of pyridine, 12 mmol of bromine, 9.5 mmol of NaNO 2, 25 ml of acetonitrile and 2.5 ml of water and then stirred for a further 15 minutes. Yield 2.21 g. Quality: relatively good. Dibromide monobromide: ethyl acetate = 13.5: 2.2: 2.3. Maximum yield 70.3%. Estimated actual yield 58-61%.

(i) Using dimethyl sulfone, 8 mmol amino-PAS, 22 mmol HBr, 12 mmol dimethyl sulfone, 12 mmol bromine, 9.5 mmol NaNO 2, 25 mL acetonitrile, and 2.5 mL water. Yield 2.50 g. Quality: at least good. Dibromide: monobromide: ethyl acetate: dimethyl sulfone = 14.5: 1.1: 1.4: 2.5. Maximum yield 76.5%. Estimated actual yield 68-70%.

(j) Using pyrrolidin-2-one 8 mmol amino-PAS, 22 mmol HBr, 12 mmol pyrrolidin-2-one, 12 mmol bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and 2.5 ml water. Yield 2.06 g. Quality: at least good. Dibromide: monobromide: ethyl acetate: pyrrolidin-2-one: water = 8: 4.6: 2: 1.4: 2. Maximum yield 66.3%. Estimated actual yield 59-61%.

Il 43 8 0 0 3 8 (k) Using guanidine 8 mmol amino-PAS, 22 mmol HBr, 12 mmol guanidine hydrochloride, 12 mmol bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and 2.5 ml water. Yield 2.08 g. Quality: at least good. Dibromodirmonobromide: ethyl acetate: water = 10.2: 3.5: 1.8: 6. Maximum yield 66.6%. Estimated actual yield 60-62%.

(l) Using dicyanomethane 8 mmol amino-PAS, 22 mmol HBr, 12 mmol dicyanomethane, bromine, 9.5 mmol NaNO 2, 25 ml acetonitrile and 2.5 ml water. Yield 2.32 g. Quality: good. Dibromide: monobromide: ethyl acetate: dicyanomethane = 9: 4.5: 2.4: 6. Maximum yield 70.8%. Estimated actual yield 62.65%.

(m) Using 1,2-dicyanoethane 8 mmol amino-PAS, 22 mmol HBr, 12 mmol 1,2-dicyanoethane, 12 mmol bromine, 9.5 mmol Na 2 Cl 2, 25 ml ethyl acetate and 2.5 ml of water. Yield 2.45 g. Quality: very good. Dibromide: monobromide: ethyl acetate: 1,2-dicyanoethane = 12.6: 1.2: 1.6: 9.7. Maximum yield 68.0%. Estimated actual yield 61-64%.

Example X

Diazotization of 68-amino-penicillanic acid 1,1-dioxide on a larger scale to 6,6-dibromo-penicillanic acid 1,1-dioxide and 6α-bromo-penicillanic acid 1,1-dioxide in the presence of 1-methylimidazole_ (a) Repetition of Example VIII (c) on a 40 mmol scale. With stirring and at a temperature below 5 ° C, 10 g (40 mmol) of 68-aminopenicillanic acid 1,1-dioxide were added to a solution of 12.5 ml of a 47% aqueous solution of hydrogen bromide (corresponding to 110 mmol of HBr) in 125 ml. acetonitrile prepared at low temperature. The starting material dissolved almost immediately. Immediately thereafter, 4.95 mL (60 mmol) of 1-methylimidazole and 2.30 mL (45 mmol) of bromine were added sequentially. As soon as the resulting mixture reached a temperature below 5 ° C by external cooling with 44 80038 using a mixture of water and crushed ice, 3.3 g (47.5 mmol) of sodium nitrite were added in approximately five equal portions over about 10 minutes and then stirring was continued for a further 30 minutes. min. While maintaining the temperature at or below 5 ° C, a solution of 5 g of sodium metabisulfite (Na 2 S 2 O 5) in 100 ml of water was carefully added, and immediately thereafter 4N sodium hydroxide was added until a pH of about 5 was reached. The resulting solution was concentrated in vacuo. 12-20 mm Hg 10-15 minutes to azeotropically remove acetonitrile. The pH of the resulting solution was raised to 6.0-6.5 with 4N sodium hydroxide and then extracted twice with a volume of about 200 ml of dichloromethane. The addition of 4N hydrochloric acid brought the resulting slightly purified solution to a pH of 2.0 and then extracted three times with 100-150 ml volumes of ethyl acetate at a pH of about 2.0. To the combined extracts was added 5 ml of saturated aqueous sodium chloride solution, and then the mixture was shaken vigorously and the layers were separated. This purification of the combined extracts was repeated with about 10 ml of saturated sodium chloride solution. The extract was dried over anhydrous magnesium sulfate, filtered, the filter cake was washed three times with a small volume of ethyl acetate and the combined filtrates were evaporated in vacuo. The product obtained was dried in vacuo.

Yield 9.83 g.

Given the presence of unknown degradation products, the quality of the separated product was relatively good as determined by TLC and PMR spectra as a 60 Mc solution in dg-DMSO and using tetramethylsilane as an internal reference. The spectrum showed a molar ratio of 6,6-dibromo-penicillanic acid 1,1-dioxide, 6α-bromo-penicillanic acid 1,1-dioxide to ethyl acetate of 8:28:18. The calculated maximum yield was 65.8%. As the quality was found to be slightly better than poor, the actual yield is assumed to be somewhat lower than 65.8% -10% of this percentage, i.e. 59.3%. The spectrum taken from the weighed amounts of reference substance and products showed that the actual yield was 55.4% of usable products.

(B) Repetition of Example VIII (b) on a 40 mmol scale. Example (a) was repeated in an identical manner but with the addition of 3.10 ml (i.e. 60 mmol) of bromine. Yield 9.98 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate = 9: 6: 1.5. Calculated maximum yield 67.8%. Since the quality of this product is significantly better than that of the product made in (a) above, the estimated actual yield is assumed to be 67.8% minus 5-10% of this value, i.e. the actual yield is 61-64%.

Comparing the results of this Example X with the corresponding results from Examples VIII (b) and (c), it appears that at least when certain tertiary amines are used as excipients, scaling seems possible even without optimization or using an even larger scale and that the approximate presentation of estimated actual yields depends entirely on , that even without precautionary measures the yields have not been corrected by taking into account the water in the starting material and the purity of the starting material.

Example XI

Diazotization Bromination in the Presence of 5,5-Dimethylhydantoin The experiments described below were performed to determine the possibility of carrying out the treatment on a larger scale using an amide-bonded excipient. 5,5-Dimethylhydantoin was chosen for this purpose because it provides reasonably high yields in diazotization bromination under conventional conditions using the preparation methods of Examples VIII and IX. The scale used in Example VIII (1) -1 was thus extended sixfold, but the remainder of Example XI is an exact copy of Example VIII (1) -1 except for one change, namely no serious attempts were made to remove the excipient, without more thorough purification of the aqueous layer after diazotization by extraction with dichloromethane. at pH 6-6.5, or by purifying the aqueous layer after reduction by extraction with, for example, ethyl acetate at a pH of about 5, which is a more suitable way to remove this excipient. The reduction step itself is one of the advantages of reducing the two-phase system, for example, in which case the actual separation and redissolution of the bromide mixture can be omitted.

While cooling with ice water, 12 g (48 mmol) of 60-aminopenicillanic acid 1,1-dioxide were dissolved with stirring in a solution of 15 ml (132 mmol) of 47% hydrobromic acid in 150 ml of acetonitrile. Continuously, 9.24 g (72 mmol) of 5,5-dimethylhydantoin and 3.72 ml (72 mmol) of bromine were added successively at 0-5 ° C, followed immediately by 3.96 g (57 mmol) of sodium nitrite. in five equal portions within 10-15 minutes. After stirring for an additional 30 minutes, a solution of 6 g of sodium metabisulfite in 60 ml of water and 4N sodium hydroxide was added with gentle stirring. After the pH reached about 6.5, the acetonitrile was removed by concentration in vacuo and then extracted twice with 60 ml of dichloromethane. The remaining aqueous solution was extracted three times at pH 2.0 with 150 ml volumes of ethyl acetate. The combined extracts were washed twice with a small volume of saturated sodium chloride solution and then mixed with 150 ml of cold water, after which the pH was adjusted to 3.5 by adding 4N sodium hydroxide.

Example XII

diazotization of β-amino-penicillanic acid 1,1-dioxide to mixtures of 6,6-dibromo-penicillanic acid 1,1-dioxide and 6α-bromopenicillanic acid 1,1-dioxide with alkyl nitrite as nitrating agent_

In the above examples, the diazotization bromination of 6g-aminopenicillanic acid 1,1-dioxide, which most likely comprises the highly water-sensitive intermediate 6-diazonium penicillanate 1,1-dioxide or its conjugate acid, was always carried out with alkali metal nitrite as the nitrating agent. The examples below show that the same type of reaction can also be carried out using alkyl or cycloalkyl nitrite as the nitrating agent. The results further show that good yields of I] 80038 can also be obtained, but that the optimized process may contain a slightly different molar ratio of starting material, acid, hydrobromic acid, bromine, excipient and water compared to a similar optimized process comprising the same excipient, but with sodium nitrite as the nitrating agent. Although various (cyclo) alkyl nitrites can be used, in the following examples only pentyl nitrite, a commercially available product consisting of a fairly accurate 3: 1 mixture of pentyl-2-nitrite and pentyl-1-nitrite, was used.

The method used on a small scale (about 8 mmol) is very similar to the method used to carry out the reaction using sodium nitrite in the above examples except for one change. Pentyl nitrite was added in one portion and then stirred for 45 minutes at 0-5 ° C. The separation treatment was also similar to that described above and always involved washing the aqueous solution with dichloromethane at pH 6.0-6.5 before extracting the product with ethyl acetate at pH 2. The experiments are therefore shown in abbreviated form in the same manner as in Examples VIII and IX.

(a) Reaction under pressure with 1-methylimidazole using 8 mmol of 6-amino-PAS, 22 mmol of HBr, 12 mmol of 1-methylimidazole, 12 mmol of bromine, 12 mmol of pentyl nitrite, 25 ml of acetonitrile and 2.5 ml of water. Yield 2.19 g. Quality: relatively good. Dibromide monobromide: ethyl acetate = 13: 2: 0.7. Maximum yield 71.2%. Estimated actual yield 61-63%.

(b) Using 1-methylimidazole 8 mmol 6-amino-PAS, 22 mmol HBr, 12 mmol 1-methylimidazole, 12 mmol bromine, 11 mmol pentyl nitrite, 30 ml acetonitrile and 2.5 ml water. Yield 2.09 g. Quality: good. Dibromide: monobromide: ethyl acetate = 12: 2: 0.8. Maximum yield 67.9%. Estimated actual yield 60-62%.

48 8 0 0 3 8 (c) Using pyridine 8 mmol amino-PAS, 28 mmol HBr, 12 mmol pyridine, 12 ramol bromine, 16 mmol pentyl nitrite, 25 ml acetonitrile, 3.2 ml water. Yield 2.65 g. Quality: only moderate. Dibromodirmonobromide: ethyl acetate = 9.5: 3: 1. Maximum yield 87.4%. Estimated actual yield: about 70% possibly more.

(d) Using pyridine 8 mmol amino-PAS, 10.5 mmol p-tolylsulfonic acid, 8.0 g (25 mmol) pyridine hydrobromide perbromide, 16 mmol pentyl nitrite, 25 mL acetonitrile, no excess water. Yield 1.92 g. Quality: only moderate. Dibromide: mono-bromide: ethyl acetate: water = 13: 1: 0.9: 5. Maximum yield 60.4%. Estimated actual yield about 50%.

Example XIII

Diazotization bromination in the presence of excess acid and using pentyl nitrite_ (a) Working continuously at 0-5 ° C, 12 g (48 mmol) of 68-aminopenicillanic acid 1,1-dioxide were dissolved in a mixture of 19.2 ml of 47% aqueous hydrogen bromide (168 mmol) and 150 mL of acetonitrile, followed by the addition of 5.7 mL (72 mmol) of pyridine and 3.72 mL (72 mmol) of bromine. A solution of 12 ml (96 mmol) of pentyl nitrite in 30 ml of acetonitrile was added dropwise over 10-15 minutes and then stirred for a further 30 minutes. The resulting reaction mixture was worked up as usual: dilute sodium metasulfite and 4N sodium hydroxide were added to pH 6.0-6.5, the acetonitrile was removed in vacuo, extracted twice with dichloromethane, the desired products were extracted with ethyl acetate at pH 2.0 and washed with ethyl acetate. by volume of sodium chloride solution.

(b) Experiment (a) was repeated using another excipient and somewhat modified conditions. Working continuously at 0-5 ° C, 12 g (48 mmol) of starting material was dissolved in the mixture,

II

49 80038 containing 9.18 g (48 mmol) of p-tolylsulfonic acid monohydrate, 15 ml of a 47% aqueous solution of hydrogen bromine (130 mmol HBr) and 150 ml of acetonitrile, followed by the addition of 6.0 ml (72 mmol) of 1 -methylimidazole and 3.72 ml (72 mmol) of bromine. 10.5 ml (84 mmol) of pentyl nitrite were added dropwise over 10-15 minutes and then stirred for a further 30 minutes. The resulting reaction mixture was treated in the usual manner as in (a). After washing with a small volume of saturated sodium chloride, the final ethyl acetate solution was dried over anhydrous magnesium sulfate, filtered and evaporated in vacuo. Yield 6.22 g. The PMR spectrum showed very good quality and that the molar ratio of 6,6-dibromo-penicillanic acid 1,1-dioxide, 6α-bromo-penicillanic acid 1,1-dioxide, ethyl acetate to pentanols was 10.6: 3 , 6: 1: 2. The calculated maximum yield was 66.5%. The estimated actual yield was 60-63%.

Example XIV

Preparation of potassium 6α-bromo-penicillanate-1,1-dioxide While maintaining the internal temperature in the range of 0-5 ° C, 20 g (80 mmol) of 68-amino-penicillanic acid 1,1-dioxide were dissolved with rapid stirring in a mixture of 300 ml of acetonitrile and 28.5 ml of 47% aqueous hydrogen bromide (corresponding to about 250 mmol of HBr), followed immediately by the addition of 15 ml (about 120 mmol) of 2-dimethylaminopyridine and 6.2 ml (120 mmol) of bromine. 6.6 g (95 mmol) of sodium nitrite were added in six equal portions over 15-20 minutes and then stirred for a further 30-40 minutes. While carefully adding 4N sodium hydroxide, the pH of the well-stirred mixture was raised to 6.5 at the same time as a solution of 10 g of sodium metabisulfite in 100 ml of cold water at pH 2-4 was added. The acetonitrile was removed in vacuo and then the volume of the aqueous solution was added to about 200 ml. The solution was extracted twice at pH 6.5 with a volume of 150 ml of dichloromethane. The combined organic phases were washed once with 50 ml of water and then discarded. 200 ml of ethyl acetate were added to the combined aqueous layers with stirring, after which the pH was adjusted to 2.0 by adding 4N hydrochloric acid. After separation of layers 80038, the aqueous layer was extracted three times at pH 2 with 150 ml volumes of ethyl acetate. 5 ml of saturated aqueous sodium chloride solution was shaken with the combined ethyl acetate extracts. After separation of the layers, the washing was repeated using 10 ml of saturated sodium chloride. Activated carbon was added to the ethyl acetate solution with stirring, and then, after 30 minutes, anhydrous sulfate was added. The mixture was filtered through a filter aid and washed with ethyl acetate. The filtrate was evaporated in vacuo to give a slightly colored solid. After drying in vacuo, the yield was 20.7 g. The PMR spectrum showed that the molar ratio of 6α-bromo-penicillanic acid 1,1-dioxide to ethyl acetate was 2: 3. The quality of the product was good and the contamination by 6,6-dibromide was negligible. The maximum yield was calculated to be 58%. The actual yield was calculated to be 51-53%.

20 g of this product was dissolved in a minimum volume of ethyl acetate, followed by the addition of a slightly excess of potassium α-ethyl capronate (approximately 5N solution in butanol). Seeding gave a crystalline product. After approximately the same volume of diethyl ether was slowly added with shaking, the product was collected by filtration, washed with cold ethyl acetate and diethyl ether. After drying in vacuo, the yield of the potassium salt of 6α-bromo-penicillanic acid-1,1-dioxide was 13 g.

Example XV

Preparation of Crude 6-Diazopenicillanic Acid-1,1-Dioxide Solutions 2.0 g of 66-aminopenicillanic acid-1,1-dioxide (containing 6-7% by weight of water) were suspended in nitrogen under stirring with any good quality acetonitrile. At a temperature of about 20 ° C with stirring, 1.5 ml (about 50% excess) of n-pentyl nitrite (Merck-Schuchardt; actually a 3: 1 mixture of pentyl-2-nitrite and pentyl-1-nitrite) was added in one portion. The reaction was observed on the basis of a temperature rise of about 3 ° and on the basis of a gradual dissolution of the poorly soluble starting material. In practice, no nitrogen evolution could be observed.

Il 5i 80038

After stirring for an additional 15 minutes, a small portion of the reaction mixture was subjected to thin layer chromatography on silica gel using 98/2 ethyl acetate / formic acid as eluent. After examination with UV light and an iodiazide / starch jet, the plate had a broad mark at an Rf value of about 0.5, and much smaller marks at the beginning and an Rf value of about 0.4, the latter presumably showing contamination with a very small amount of no. -desired by-product, namely 3,3-dimethyl-8-oxo-1-aza-4-thia-7-oxa-bicyclo [3.30] octane-2-carboxylic acid-4,4-dioxide. A small volume of the reaction mixture diluted with acetonitrile was subjected to IR spectroscopy using acetonitrile for comparison. The obtained spectrum had a remarkably strong absorption at 2150 cm caused by the diazo group.

1) TLC and IR regulations can therefore show a very good, possibly about 90% conversion of the starting material to 6-diazopenicillanic acid 1,1-dioxide despite the presence of water in the starting material and the solvent which had not been dried.

2) Suspensions of the starting material among other Good quality, but not substantially dried, non-hydroxyl-containing solvents, such as in tetrahydrofuran and other ether-type solvents, methyl acetate and ethyl acetate, were also suitable for reaction at room temperature with alkyl nitrite to give solutions of penicillanic acid 1,1-dioxide of approximately the same quality. Due to too low solubility or reactions of the diazo group with the solvent, unsuitable solvents included halogenated hydrocarbon solvents such as dichloromethane (too low solubility), a few hydroxyl-containing solvents such as ethanol, and solvents containing a reactive group such as a reactive carbonyl group. in experiments using ethanol, considerable dissolution of the starting material occurred, but was followed by significant degradation or alteration of 6-diazopenicillanic acid 1,1-dioxide.

52 80038 3) The above-mentioned bicyclo [3.3.0] octane-2-carboxylic acid 4,4-dioxide may be present as a minor impurity due to the Rf value, which is identical to the Rf value of the authentic compound, and since several times that when nitrating the starting material under different acidity conditions of the intermediate, i.e. between about pH 2 and pH 5, possibly in the presence of bromide anions, but especially in the absence of bromine, this product is very likely, if not certainly, present in the reaction product mixture as a result of reaction of water with a nitrated medium but not a diazo group or diazonium group. As a result of such a reaction mechanism, an oxygen atom of a water molecule is placed between the original carbon atom 6 and the carbon atom 7 of the bicyclic group. The best established way to prepare such a bicyclic compound is to add sodium nitrite to a solution of 66-aminopenicillanic acid 1,1-dioxide in water prepared by dissolving the triethylamine salt of the starting material at pH 4.5-5.0. Since under these conditions the conversion of the starting material to the bicyclo compound is about 80% or more, and since the extraction with ethyl acetate is simple, the compound can be easily obtained and thus in pure form. The identity of the compound, 3,3-dimethyl-8-oxo-1-aza-4-thia-7-oxa-bicyclo [3.3.0] octane-2-carboxylic acid-4,4-dioxide, was established by mass spectrogram of its methyl ester derivative. min, which was obtained by reaction with diazomethane, which confirmed the molecular weight of the ester to be 249, and the IR and NMR spectra of this stable, hitherto unknown bicyclic sulfonic carboxylic acid (after conversion to the potassium salt) (KBr plate values in cm). 1): 3250, 3200, 1758 (very powerful), 1615 (very powerful), 1452, 1395 (powerful), 1342 (powerful), 1308 (powerful), 1260, 1200 (very powerful; shoulders), 1180 (sh) , 1140, 1105 (strong), 1080, 1035, 988, etc. PMR (dg-DMSO and a few drops DCX ^ D, TMS, 60 Mc, δ values in ppm): 1.43 (s, 3H), 1 .52 (s, 3H), 4.45 (s, 1H), 4.62-5.34 (ABC decomposed pattern, C5-H multiband at about 5.25, 3H).

Il 53 80038 4) Notwithstanding the hydroxyl compounds, i. in the presence of small but about equimolar amounts of water and pentanol, such about 90% pure solutions of 6-diazopenicillanic acid 1,1-dioxide are surprisingly stable at or below 0 ° C and can therefore be stored. Since complete decomposition during storage at room temperature, which requires several days to complete, as well as at 50 ° C, which requires several hours, resulted in a product mixture in which another new compound is predominant, decomposition at 50 ° C was also performed in the presence of about 12 mol% excess of ethyl tetrolate (1-methyl-2-carbethoxyethine). The decomposition was now complete in about 30 minutes and the new compound, which is also inherently unstable, is present in the reaction product in a much higher amount. This compound was isolated in about 40% yield as a non-completely pure product by collecting the precipitates after several concentrations in vacuo, which were combined while cooling and diluting with diethyl ether, as well as by lyophilization. Starting from 2.0 g of 60-amino-penicillanic acid 1,1-dioxide, 1.19 g of a partially crystalline product were obtained. The decomposition was accelerated because this new compound was first assumed to be 5,5-dimethyl-4,5-dihydrothiazole-3-carboxylic acid 1,1-dioxide, causing thermal decomposition to this compound and a hypothetical diazoketone, which could then be isolated by ethyl tetrolate. However, this structure had to be rejected for several reasons. This new compound was formed in the reaction between water and the 6-diazo derivative to give nitrogen and was promoted by the presence of ethyl tetrolate. The IR spectrum of the product (KBr plate values in cm was quite ambiguous due to the clear hygroscopic nature of the product. about 1290, about 1240 and 1070. The spectrum showed the presence of considerable water Very good quality compared to other sulfur-containing compounds, determined by TLC, was confirmed by PMR spectrum (d-DMSO, 6 values

O

ppm, TMS, 60 Mc) disregarding the presence of about 1/3 mole of acetic acid and about 2 moles of water per mole of compound: C (CH 3) 2: 1.04 (s, 3H) and 1.33 (s, 3H): C5-H: 54,80038 5.69 (s, 1H) and C2-H: 8.82 (s, 1H). The spectrum showed the presence of 4-6 mole% of the second compound, which also had two CH-j simple signals and two C-H simple signals, probably from the regrouping of the main product. Based on the distribution of absorption signals in the PMR spectrum, it was hypothesized that this new compound was 6,6-dimethyl-5,6-dihydro- (4H) -thiazine-3,5-dicarboxylic acid 1,1-dioxide, which the formation mechanism may correspond to. Further identification of the structure, for example one which is possible by conversion to a monoester or diester using a diazane reaction, was not carried out because this compound, as well as 3,3-dimethyl-8-oxo-1-aza-4-thia-7- oxa-bicyclo- [3,3,0] octane-2-carboxylic acid-4,4-dioxide, has no antibiotic properties. These two novel compounds have been described herein to further identify 6-diazopenicillanic acid-1,1-dioxide, which constitutes an important embodiment of the present invention, and to describe the difficulties which had to be overcome in order to obtain the 68-aminopenicillanic acid-1 according to the invention. Suitable diazotization methods for 1-dioxide.

Example XVI

Preparation of crude potassium 6-diazopenicillanate-1,1-dioxide 2.0 g of 68-aminopenicillanic acid 1,1-dioxide were suspended in 25 ml of methyl acetate with stirring. The reaction with pentyl nitrite was carried out as described in Example XIV. After completion of the reaction, which was adjusted by TLC, about 1 molar n-butanol solution of potassium α-ethyl capronate was gradually added at 10 ° C until the pH of the mixture was 6.5. Since no precipitation occurred, 50 ml of anhydrous diethyl ether was added dropwise with stirring. After standing for 30 minutes at 0-5 ° C, the precipitate was collected by filtration, washed rapidly with cold methyl acetate and then with diethyl ether. The hydroscopic yellowish solid was dried in vacuo in a desiccator over several hours. Yield 2.2 g. This

II

55 80038 The new potassium salt of 6-diazo-penicillanic acid-1,1-dioxide thus obtained, which appeared to be relatively stable when stored in the refrigerator, was analyzed and examined by TLC, IR and PMR analyzes.

TLC: A small amount of product was dissolved in ice-cold water.

The solution was subjected directly to TLC analysis on silica using 98: 2 ethyl acetate / formic acid. Examination under UV light after drying the plate and in combination with iodiazide / starch after heating to 150 ° C showed non-negligible sites at the beginning (one or more degradation products) and one major site at an RF value of about 0.5 (desired product). The solution was allowed to stand at about 18 ° C for about 1 hour. TLC analysis on silica using acetonitrile: water: formic acid 80: 20: 1 showed strong but not complete decomposition to thiazine-1,1-dioxide (Rf 0.24; visible under UV light, almost imperceptible on iodine azide / of starch). This conversion of 6-diazopenicillanic acid-1,1-dioxide (Rf is 0.72 in this system) occurred in an almost selective manner. Significant formation of 3,3-dimethyl-8-oxo-1-aza-4-thia-7-oxa-bicyclo [3.3.1] octane-2-carboxylic acid 4,4-dioxide could not be detected on these plates. . IR: (KBr plate values in cm x): very strong absorptions at 1780 (8-lactam) and 1630 (carboxylate anion), strong absorptions at 2155 (diazo group) and 1335 (SO2), slightly less strong absorptions at 1400 , 1265, 1170, and 1135, and lower absorbances at about 3000, 1485, 1060, 820, and several others between 1040 and 640. The spectrum also indicated the presence of water.

PMR: (dg-DMSO δ values in ppm, TMS, 60 Mc): The spectrum showed the presence of about 1 mole of water per mole of compound and a small impurity of acetic acid, methyl acetate, diethyl ether and α-ethyl capronate due to the crystalline nature of the product, but no decomposition products actually occurred. 6-Diazopenicillanic acid 1,1-dioxide had the following absorptions: C (CH 3) 2: 1.36 (s, 3H), 1.46 (s, 3H); C3-H: 3.66 (s, 1H) and C8-H: 5.81 (s, 1H).

56 80038

The addition of a small amount of DCO 2 D 2 resulted in a considerable decomposition of the thiazine-1,1-dioxide compound. At 25 ° C, a solution of the product in D 2 O gave a spectrum showing immediate and continuous decomposition to the same substance in an almost pure reaction, which was also detectable by TLC.

(a) A careful comparison of the integrals of the absorptions belonging to water and the various impurities mentioned above with the absorption integrals due to the diiso derivative gave a calculated yield of 1.67 g instead of 2.22 g, i.e. a separation yield of just over 70%. In practice, this result shows a very good conversion of 6β-penicillanic acid-1,1-dioxide to 6-diazopenicillanic acid-1,1-dioxide as observed by TLC in Example XIV.

b) The corresponding crude sodium salt was prepared in a similar manner and in approximately the same yield from an in situ solution of ethyl acetate. Drying conditions during the preparation of the 6-diazo derivative and its salts did not provide substantially better yields. In the absence of a different acid, 6-diazopenicillanic acid 1,1-dioxide and its salts apparently remain reasonably stable in the presence of a small amount of water in the pH range of about 2-7.

Example XVII

Conversion of 6β-amino-penicillanic acid-1,1-dioxide to 6,6-di-bromoxyn- and 6a-bromo-penicillanic acid-1,1-dioxides by pyridine hydrobromide perbromide and alkyl nitrite and comprising 6-diazo-penicillanic acid-1 In the experiments described below, pentyl nitrite was added at 15-20 ° C to a suspension of 6 β-aminopenicillanic acid 1,1-dioxide in an organic solvent, followed by stirring for about 10 minutes under nitrogen until complete dissolution of the starting material. had happened. The solution was cooled to 5 ° C, followed by the addition of the bromination complex in one portion, and stirring for about 30 minutes until TLC indicated complete reaction of 6-diazopenicillanic acid 1,1-dioxide. As repeatedly described above, the reaction mixture was treated with dilute aqueous sodium metabisulfite and 4N sodium hydroxide. The organic solvent was removed in vacuo at pH 6-6.5. With the exception of experiment (a), in which the bromide mixture was then directly extracted at pH 2 with ethyl acetate, etc., in other experiments, the aqueous solution was first purified by extraction twice with dichloromethane at a pH of about 6.5. The experiments are presented in abbreviated form, which was already used in Examples VIII and IX.

(a) 8 mmol of amino-PAS, 11 mmol of commercially available pentyl nitrite, 12 mmol of pyridine hydrobromide perbromide (Cj-Hj-N.HBr ^), 20 ml of acetonitrile. Yield 2.28 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate: pentanol = 6: 5: 1.3: 6.6. Maximum yield 68.8%, disregarding the amount of water in the starting material. Estimated actual yield 62-64%.

The PMR spectrum (D 6 -DMSO) (δ values in ppm, TMS, 60 Mc) for 6α-bromopenicillanic acid 1,1-dioxide and 6,6-dibromo-penicillanic acid 1,1-dioxide had the following absorption signals.

6α-bromide: C (CH3) 2: 1.41 (s) and 1.51 (s), C3-H: 4.50 * C8-H and C8-H: about 5.6 (narrow AB quartet, low chemical transfer difference, characterized by a low Jg6 value for trans-orientation).

6,6-dibromide: C (CH 3) 2: 1.41 (s) and 1.51 (s) (complete agreement with the corresponding signals of 6α-monobromide) C 3 H: 4.68 (s); C5-H: 6.00 (s).

(b) Experiment (a) was repeated using 20 ml of ethyl acetate instead of 20 ml of acetonitrile. Yield 2.02 g. Quality: from good to very good. Dibromide: monobromide: ethyl acetate = 2: 7: 7.5. Maximum yield 62.7%. Estimated actual yield 56-59%.

58 80038 (c) 8 mmol of amino-PAS, 11 mmol of pentyl nitrite, 25 mmol of pyridine hydrobromide (no bromine added), 20 ml of acetonitrile. By TLC, the addition of pyridine.HBr to the 6-diazo derivative formed in situ did not result in significant formation of useful products.

(d) 8 mmol of amino-PAS, 11 mmol of pentyl nitrite, 8 mmol of pyridine.HBr, 8 mmol of pyridine hydrobromide perbromide, 20 ml of acetonitrile. Yield only 1.11 g.

(e) 8 mmol amino-PAS, 11 mmol pentyl nitrite, 24 mmol pyridine hydrobromide perbromide, 20 ml acetonitrile. Yield 0.93 g.

(f) 8 mmol amino-PAS, 11 mmol pentyl nitrite, 12 mmol pyridine, 12 mmol bromine (no hydrobromic acid), 20 ml acetonitrile. Yield only 1.26 g.

In the absence of hydrobromic acid, since the reaction at 5 ° C was considerably slower as determined by TLC analysis, the temperature was allowed to rise to 20 ° C within 10 minutes after the addition of bromine. The reaction was complete about 40 minutes after the addition of bromine. The relatively low yield indicated that it is necessary to use bromine as well as hydrobromic acid. The same experiment using 1-methylimidazole instead of pyridine gave a yield of only 0.95 g.

Example XVIII

Preparation of a mixture of pivaloyloxymethyl esters of 6,6-dibromo and 6a-bromo-penicillanic acid 1,1-dioxide_ a) Preparation of pivaloyloxymethyl-6a-amino-penicillanic acid 1,1-dioxide

A mixture of 14.84 g (0.144 mmol) of sodium bromide, 17.5 ml (0.121 mmol) of pivaloyloxymethyl chloride and 320 ml of dry methylformamide was stirred at room temperature for 45 minutes. To the mixture was added 41.92 g (0.120 mmol) of the triethylamine salt of 66-aminopenicillanic acid 1,1-dioxide, followed by stirring for 3.5 h. The reaction mixture was poured.

II

59 80038 then to a well-stirred mixture of 3.2 liters of water and crushed ice, 160 g of sodium chloride and 1.6 liters of ethyl acetate, and then 4N sodium hydroxide was added until the pH of the mixture was 6.5. The layers were separated and the aqueous layer was extracted three times with 400 ml volumes of ethyl acetate at pH 6.5. The combined organic layers were washed three times with 400 ml of ice-cold water and then stirred in the cold with anhydrous magnesium sulfate for 30 min. The salt was removed by filtration, after which the filtrate was evaporated in vacuo. The residual oil was mixed with 800 ml of diethyl ether, after which the insoluble material was removed by filtration. The filtrate was diluted with 800 ml of n-hexane. The diethyl ether was removed from the cloudy solution by concentration in vacuo to precipitate the desired product as an oil. The n-hexane was then removed by decantation and then 200 ml of diethyl ether and 600 ml of n-hexane were added successively. After removal of the diethyl ether in vacuo, the semi-solid product was separated from n-hexane by decantation. The solid was shaken with n-hexane, followed by decantation of the n-hexane and drying in vacuo to constant weight. Yield 19.6 g (about 45%) of almost pure product based on thin layer chromatography and PMR spectrum.

PMR (CDCl 3, 60 Mc, δ values in ppm, TMS): 1.22 (s, 9H), 1.40 (s, 3H), 1.55 (s, 3H), about 2.4 (br , about 2H), 4.48 (s, 1H), about 4.68-4.90 (AB-q, minor br-lines, J = 4.5 Hz, 2H) and 5.65-6.00 ( AB-q, J = 5.5 Hz, 2H).

The triethylamine salt of the starting compound can be prepared in high yield as follows: 39.8 g of 68-aminopenicillanic acid 1,1-dioxide containing about 6% by weight (corresponding to about 0.75 mol%) of water were suspended with stirring for 700 with dichloromethane and then 29.7 ml of triethylamine were added dropwise. After almost complete dissolution, a small amount of insoluble material was removed by filtration. 700 ml of ethyl acetate was added to the filtrate and then concentrated in vacuo to a small volume.

60 80038

After the addition of 500 ml of ethyl acetate, the concentration was repeated in vacuo. At the same time, under ice-cooling, a small amount of diethyl ether was added while stirring or scraping to an almost oily residue. The resulting crystalline mass was passed through a glass filter, suction filtered, washed once with a small amount of diethyl ether and dried in vacuo. Yield 50 g or more of pure, slightly hygroscopic product (must be stored in a refrigerator). This salt is freely soluble in water and in a number of conventionally used organic solvents.

(b) Preparation of a mixture of pivaloyloxymethyl esters of 6,6-dibromo and 6α-bromo-penicillanic acid 1,1-dioxide Under nitrogen and while cooling in an ice bath, 10.00 g (31.25 mmol) of pyridine hydrobromide perbromide (C (H.HBr-j) to a solution of 9.05 g (25 mmol) of the ester prepared in (a) in 62.5 ml of acetonitrile. At a temperature not exceeding 8 ° C, 3.75 ml of pentyl nitrite were added dropwise. After about 1 hour, when gas evolution had completely stopped and the temperature had dropped to about 3 ° C, 3 ml of pyridine (37 mmol) was added slowly. With 50 ml of cold ethyl acetate, the contents of the reaction vessel were transferred to an evaporating flask and then a cold solution of 2.5 g of sodium metabisulfite in 25 ml of water was added with shaking. The organic solvent was quickly removed by evaporation in vacuo at low temperature. The aqueous solution was extracted with 300 ml of diethyl ether, after which the organic phase was washed three times with 25 ml of ice water. After drying over magnesium sulfate, the diethyl ether solution was evaporated to a red oil. The product was dissolved in a small volume of dichloromethane and then chromatographed on a short column of silica. After eluting with dichloromethane, the pure and almost pure fractions were combined by TLC and evaporated in vacuo. The oily, only slightly colored residue was dissolved in 100 ml of diethyl ether and then 200 ml of n-hexane were added and concentrated in vacuo to a volume of about 80 ml, whereupon the semi-solid product precipitated. After that I! 61 80038 was removed by decantation of n-hexane and triturated with a small volume of n-hexane and decanted (twice). After thorough drying in vacuo, the yield was 6.61 g. The purity of the product was 90-95% as determined by the PMR spectrum. It consisted of a 2: 1 mixture of approximately 6,6-dibromide and 6α-bromide in a yield of approximately 55%.

PMR (CDCl 3/60 Mc, δ values in ppm, TMS): 1.22 (s, 9H), 1.43 (s, 3H), 1.59 (s, 3H), 4.43 (s, 1 / 3H), 4.52 (s, 2 / 3H), 4.69 (d, J = 1.5 Hz, 1 / 3H), 5.02 (s, 2 / 3H), 5.16 (d , J = 1.5 Hz, 1 / 3H), 5.68-6.01 (AB-q, J = 5.3 Hz, 2H).

Example XIX

Preparation of pivaloyloxymethyl 6-diazo-penicillanate-1,1-dioxide and pivaloyloxymethyl-6a-bromo-penicillanate-1,1-dioxide________ (a) Crude pivaloyloxymethyl-6-diazo-penicillanate-1,1-dioxide Preparation 1.45 g (4 mmol) of pivaloyloxymethyl-68-amino-penicillanate t-1,1-dioxide prepared as described in Example XVIII (a) were dissolved in 15 ml of dry acetonitrile. Under a nitrogen atmosphere while stirring at 10 ° C, 150 ml (1.67 mmol) of anhydrous oxalic acid and 0.75 ml (5.5 mmol) of pentyl nitrite were added successively. The mixture was stirred at 10 ° C for 1 hour and then the solvent was quickly removed in vacuo and the residue was dissolved in about 25 ml of dichloromethane. TLC analysis (silica, diethyl ether - ca. hexane) indicated that a virtually complete reaction had been achieved. This solution was stored overnight at -15 ° C. The precipitate (a small amount of oxalic acid) was removed by filtration through a glass funnel. The filtrate was cooled to below -10 ° C and then washed twice with 8 ml volumes of aqueous sodium chloride at a temperature between -10 and -15 ° C. The solution was kept cold with anhydrous magnesium sulfate with occasional shaking at -15 ° C for 1 hour. The salt was removed by filtration and washed with a small volume of cold dichloromethane. The combined filtrates were evaporated in vacuo to a semi-solid, partially crystalline residue.

62 80038 The residue was triturated with n-hexane in the cold and then stored at -15 ° C for 2 hours. Now the completely solid and crystalline product was collected by filtration, washed with a cold 1: 1 mixture of n-hexane and carbon tetrachloride and then dried thoroughly in vacuo. Yield 1.2 g of about 80% pure product.

IR (KBr plate, values in cm 1): mm. absorptions at wavelengths 2980 (m), 2135 (m), 1775 (s), 1755 (vs), 1330 (s), 1120 (s).

PMR (CDCl 3) / 60 Mc, δ values in ppm, TMS): 1.23 (s, 9H), 1.43 (s, 3H), 1.58 (s, 3H), 5.65- 5.98 (AB-q, J about 6.0 cps, 2H), 4.24 (s, 1H) and 5.47 (s, 1H).

(b) Preparation of pivaloyloxymethyl 6a-bromo-penicillanate-1,1-dioxide 1.28 g (4 mmol) of pyridine hydrobromide perbromide were added portionwise over 10 minutes to a stirred solution of 1 at 1-5 ° C. 1 g of the crude diazo derivative prepared in (a) above in 10 ml of dry acetonitrile. The ice bath was removed and the solution was allowed to reach room temperature (about 20 ° C) and then stirred for an additional 10 minutes. TLC analysis (silica, 2: 1 mixture of diethyl ether and n-hexane) showed that the diazo derivative was completely reacted. Then, 0.4 ml of pyridine and a solution of 0.4 g of sodium metabisulfite in 4 ml of water were added successively under cooling. The acetonitrile was removed by concentration in vacuo, after which the remaining mixture was extracted three times with about 12 ml of diethyl ether. The combined extracts were washed with saturated aqueous sodium chloride solution, then water, dried over anhydrous magnesium sulfate, filtered and evaporated in vacuo. The residue was dissolved in a small volume of dichloromethane and the solution was chromatographed using a short column of silica and dichloromethane as eluent. Yield 0.9 g of almost pure pivaloyloxymethyl 6a-bromo-penicillanate t-1,1-dioxide. The presence of 6,6-dibromide could not be detected by the PMR spectrum.

Il 80038 63

Example XX

Preparation of pivaloyloxymethyl 6,6-dibromo-penicillanate-1,1-dioxide_______

Diazotization was performed as described in Example XIX starting from 2.9 g (8 mmol) of pivaloyloxymethyl-66-aminopenicillanate-1/1-dioxide and using 360 mg (4 mmol) of oxalic acid, 1.5 mL (11 mmol) of pentyl nitrite and 15 ml of acetonitrile. After complete reaction of the starting material, 3.84 g (12 mmol) of pyridinium hydrobromide perbromide was added as described in Example XIX. After the complete reaction, 1 ml of pyridine and a solution of 1 g of sodium metabisulfite were added successively. The acetonitrile was removed in vacuo and the remaining mixture was extracted three times with 30 ml volumes of diethyl ether. The crude product was subjected to column chromatography as described in Example XIX. Yield 1.92 g of almost pure pivaloyloxymethyl 6,6-dibromo-penicillanate-1,1-dioxide based on the PMR spectrum. The final product contained about 5 mole percent unknown by-product. The corresponding 6α-mono-bromide was not present based on the PMR spectrum.

Example XXI

Preparation of crude mixtures of 6,6-dibromo-penicillanic acid 1,1-dioxide and 6α-penicillanic acid 1,1-dioxide by varying the reactants and excipients_ HBr was added as a 47% solution to the stirred and cooled (0-5 ° C) suspension. with APZ sulfone in acetonitrile. A clear solution was obtained. The excipient and bromine were then added to the above solution, followed by portionwise addition of sodium nitrite over 15 minutes, maintaining the temperature between 0 and 5 ° C. The resulting mixture was then stirred for an additional 30 min (0-5 ° C) and sodium metasulfite solution was added so that the temperature did not exceed 5 ° C. The pH was then raised to 5.5 with 1N NaOH and the acetonitrile was removed under reduced pressure (water pump). The resulting solution was adjusted to pH 6.5 with 4N NaOH and extracted with dichloromethane. The aqueous layer was then adjusted to pH 2 with 4N HCl and extracted 64,80038 three times with ethyl acetate. The combined ethyl acetate extracts were dried over MgSO 4 and the solvent was removed under reduced pressure and the product was then dried over P 2 O 4 under reduced pressure and weighed (yields are shown in grams in Table 1).

The method was changed as described in Table A.

Table A

Example XXI - Experiments a-n

Eg HBr, Amino-PAS, Br ~ Nitrite, Acetonitrile, Excipient itmoles niroles mmol_ml_ a 22 8 12 9.5 NaNCL 25 12 mmol 4-methylpyrimidine b 22 8 12 9.5 NaNCL 25 0.8 itmoles 4 -methyl-pyrimidine c 22 8 12 9.5 NaNO 2 25 0.8 mmol caprolactam d 220 80 120 95 NalCt 250 8 itmoles caprolactam e 22 8 12 9.5 NaNO 25 0.8 itmoles N-vi - Nylimidazole f 110 40 12 47.5 NaNO 2 125 4 itmoles N-vinylimidazole g 22 8 12 9.5 NaNO 2 25 0.8 itmoles succinimide h 110 40 60 47.4 NaNO 25 4 itmoles N-benzyl-benzenesulfonamide 22 8 12 9.5 NaNO 2 25 0.8 itmoles 2-dimethylaminopyridine and 22 8 12 9.5 NaNO 2 25 0.8 itmoles ethyl cyanoacetate k 110 40 60 47.5 NaNO 2 25 0.45 itmoles ethyl cyanoacetate 1 22 8 12 12 penyl 25 0.8 itmoles methylnitrite liimidazole m 22 8 12 11 penyl 25 0.8 itmoles methylnitrite imidazole n 28 8 12 16 pentyl-0.8 0.8 moles of pyrinitrite dine

The results of the above experiments are shown in Table B.

Il 65 80038

Table B

Example XXI - Yields of Examples a-n and results of NMR analyzes. Ratio = molar ratio of 6,6-dibromo-penicillanic acid to 1,1-dioxide: 6g-bromo-penicillanic acid-1,1-dioxide

Experiment Yield Ratio Ethyl acetate Maximum yield Estimated true _g_mole -% _% _ yield,% a 2.2 9: 7.5 11 74.5 69-72 b 2.42 8: 1 11 76.4 70.3 c 2 , 36 3: 1 12 76 68-70 d 21.61 10:75 10 72.4 66.5 e 2.42 13: 2 14 75.4 67.9 f 11.05 10: 7.5 13 74, 4 68 g 2.44 14: 3 12 79.5 71.5 h 11.12 11: 6.5 12 73.9 66.5 i 2.48 18: 2.5 10 78.9 71 j 2.2 1: 1 14 73.4 68-71 k 11.47 5: 3 13 75.2 67.6 1 2.40 16: 1.5 12 75.3 69.7 m 2.35 16: 1 12 73, 8 66.4 n 2.37 29: 3 13 74.3 66

A small number of unknown products were found in all experiments.

The experiments of this Example XXI show that mixtures of 6,6-dibromo-penicillanic acid 1,1-dioxide and 6a-bromo-penicillanic acid 1,1-dioxide, which are valuable intermediates in the preparation of penicillanic acid 1,1-dioxide , is obtained in very good yield when using the method according to the invention.

Example XXII

Diazotization / bromination of 68-amino-penicillanic acid 1,1-dioxide without aid using sodium nitrite as diazotization reagent __ 7.9 ml of aqueous HBr (68.75 mmol) was added to the stirred and cooled (0-5 ° C) APZ sulfone suspension. (6.2 g; purity by HPLC 91%; 22.75 mmol) in acetonitrite (75 mL).

Bromine 66 80038 (1.92 ml; 37.5 mmol) was then added to the above solution, followed by portionwise addition of sodium nitrite (2.05 g; 29.7 mmol) over 15 minutes while maintaining the temperature in the range of 0-5 ° C. The resulting mixture was stirred for an additional 30 min (0-5 ° C) and then sodium metabisulfite solution (3.1 g; 60 mL in water) was added so that the temperature did not exceed 5 C. The pH was then raised to 3 with 4N NaOH. 5 and acetonitrile were evaporated under reduced pressure (water pump) and the resulting mixture was extracted with ethyl acetate (3 x 100 ml). The combined ethyl acetate extracts were washed with brine (2 x 50 ml), dried over anhydrous magnesium sulphate and the solvent evaporated under reduced pressure to give 7.57 g of 6,6-dibromopenicillanic acid 1,1-dioxide and 6a-bromopenicillanic acid-1,1-dioxide : 6,6-dibromopenicillanic acid 1,1-dioxide: 80.7% and 6α-bromopenicillanic acid 1,1-dioxide: 7.4%) to give 6,6-dibromopenicillanic acid 1,1-dioxide 68.7 % and 6a-bromo-penicillanic acid-1,1-dioxide 7.9%. Purity was determined by 60 MHz spectroscopy using maleic acid for comparison.

Example XXIII

Diazotization / bromination of 66-amino-penicillanic acid 1,1-dioxide using auxiliary reagent and pentyl nitrite as diazotization reagent__________

While cooling with ice water, APZ sulfone (10 g; HPLC purity 91%; 44.8 mmol) was dissolved in a solution of hydrobromic acid (12.5 mL; 110 mmol) in acetonitrile (125 mL) with stirring. With continuous operation at 0-5 ° C, 1-methylimidazole (0.32 mL; 4 mmol) and bromine (3.1 mL; 60 mmol) were added sequentially, followed immediately by pentyl nitrite (8 mL; 60 mmol). After stirring for an additional 45 minutes, a solution of sodium metabisulfite (5 g) in 100 mL of water and 4N NaOH was added with stirring until the pH was 6.5. The acetonitrile was then removed by concentration in vacuo, followed by extraction twice with 50 ml of dichloromethane. The remaining aqueous solution was extracted at pH 3.5 with ethyl acetate (3 x 100 mL). The combined extracts were washed twice with a small volume of brine, dried at 67,80038 anhydrous magnesium sulfate, and the solvent was removed under reduced pressure to give 11.16 g of 6,6-dibromopenicillanic acid 1,1-dioxide and 6α-bromopenicillanic acid 1,1-dioxide : 6,6-dibromopenicillanic acid 1,1-dioxide 76.3% and 6α-bromopenicillanic acid 1,1-dioxide 11.1%) to give 6,6-dibromopenicillanic acid 1,1-dioxide in 59% yield and 6.8% yield of 6α-bromopenicillanic acid-1,1-dioxide. Purity was determined by 60 MHz spectroscopy using maleic acid for comparison.

Example XXIV

Diazotization / bromination of 68-amino-penicillanic acid-1,1-dioxide without auxiliary reagent using pentyl nitrite as diazotization reagent___

While cooling with ice water, APZ sulfone (10 g; purity by HPLC analysis 91%; 44.8 mmol) was dissolved in a solution of hydrobromic acid (12.5 mL; 110 mmol) in acetonitrile (125 mL) with stirring. Bromine (3.1 mL; 60 mmol) was added continuously at 0-5 ° C followed immediately by pentyl nitrite (8 mL; 60 mmol). After stirring for an additional 45 minutes, a solution of sodium metabisulfite (5 g in 100 mL of water) and 4N NaOH was added with stirring until the pH rose to 3.5, and the acetonitrile was removed under reduced pressure and then extracted with ethyl acetate (3 g). x 100 ml). The combined extracts were washed twice with a small volume of brine, dried over anhydrous magnesium sulfate, and the solvent was removed under reduced pressure to give 12.2 g of 6,6-dibromopenicillanic acid 1,1-dioxide and 6a-bromopenicillanic acid 1,1-dioxide (purity: 6, 6-dibromopenicillanic acid 1,1-dioxide 74.4% and 6a-bromopenicillanic acid 1,1-dioxide 9.3%) to give 63% of 6,6-dibromopenicillanic acid 1,1-dioxide and 9.9% 6a-bromopenicillanic-l, 1-dioxide. Purity was determined by 60 MHz spectroscopy using maleic acid for comparison.

68 80038

Example XXV

Example XXIV was repeated using 4.6 mL of aqueous HBr (39 mmol) and 3.3 mL of 12 NH 2 SO 4 (20 mmol) instead of 7.9 mL of HBr. Isolated: 6.06 g of product (purity: 6,6-dibromopenicillanic acid 1,1-dioxide 64.2% and 6α-bromopenicillanic acid 1,1-dioxide 26.0%). The yield was thus 44.3% 6,6-dibromopenicillanic acid 1,1-dioxide and 22.5% 6a-bromopenicillanic acid 1,1-dioxide.

Il

Claims (22)

69 80058 A process for the preparation of 6α-bromo and / or 6,6-dibromo-penicillanic acid 1,1-dioxides of the general formulas 0 0 \ / Br HS CHaX! \ / \ / H - C - C Cv I I I l \, C - N - CH CHa / O xCOOR and / or O O \ / Br HS. CHaX i \ / \ / Br - C - CC II II l \, C - N - CH CH3 / O COOR where X is hydrogen, a chlorine atom or an acetoxy radical and R is hydrogen or a metal or ester radical, characterized in that the δβ- aminopenicillanic acid-1,1-dioxide derivative of the general formula OO \ / H HS .CHaX '! / \ / HaN - C - C C. III II! C, N - CH CHa / O 'COOR wherein X and R are as defined above, followed by bromination of the diazotized compound 70 80038 with at least an equimolar amount of nitrating agent in the presence of 1-5 equivalents of strong inorganic or organic acid in solution or suspension consisting of a mixture of water and a partially or completely water-miscible organic solvent medium in an amount of 1 to 20% by volume, containing at least equimolar amounts of hydrogen bromide and bromine and optionally an excipient, or at least an equimolar amount of alkyl nitrite or cycloalkyl nitrite in solution or suspension in a hydroxyl-containing, inert, organic solvent medium containing no added water or not more than 10% by weight of water to form a 6-diazopenicillanic acid 1,1-dioxide derivative of the general formula NH = .-- Nv -.- CC CXIII ' 'CH -N-CH 30' CO-OR wherein X and R are as defined above, optionally isolating said diazo compound from the reaction mixture as such, or - when R is hydrogen - as a salt, converting the diazo compound to the general formula I and / or II 6a-bromo - and / or 6,6-dibromopenicillanic acid to the 1,1-dioxide derivative or an ester thereof by adding a complex or mixture containing at least an equimolar amount of bromine and hydrogen bromide and preferably also a tertiary amine. Il 71 80038 Process according to Claim 1, characterized in that the reaction is carried out at a temperature between -20 ° C and + 30 ° C. Process according to Claims 1 and 2, characterized in that the reaction is carried out at a temperature between -10 and + 15 ° C. Process according to Claims 1 to 3, characterized in that the strong inorganic or organic acid is arylsulfonic acid, sulfuric acid or hydrobromic acid. Process according to Claims 1 to 4, characterized in that the organic solvent medium is acetonitrile, 1,2-dimethoxyethane, ethyl acetate, methyl acetate, nitromethane or tetrahydrofuran, or a mixture of two solvents containing at least 80% by volume of acetonitrile, ethyl acetate or dichloromethane, and up to 20% by volume of sulfolane, dimethyl sulfoxide, lower alkyl cyanoacetate or di (lower) alkyl malonate. Process according to Claims 1 to 5, characterized in that the organic solvent is acetonitrile or 1,2-dimethoxyethane. Process according to Claims 1 to 5, characterized in that the organic solvent medium is a mixture of acetonitrile or ethyl acetate with ethyl cyanoacetate or diethyl malonate. Process according to one of Claims 1 to 7, characterized in that the nitrating agent used is sodium nitrite, 1-pentyl nitrite, 2-pentyl nitrite or a commercial mixture of 1-pentyl nitrite and 2-pentyl nitrite. 72 80038 Process according to Claim 8, characterized in that the nitrating agent is sodium nitrite or a commercial mixture of 1-pentyl nitrite and 2-pentyl nitrite. Process according to one of Claims 1 to 9, characterized in that X in the general formula I is hydrogen. Process according to one of Claims 1 to 10, characterized in that 6,6-dibromopenicillanic acid 1,1-dioxide is prepared using pyridine, sulfolane, hydantoin, phenylacetylurea, p-tolylsulphonamide, succinimide, diphenylsulphoxide, diphenylsulphoxide, 1-vinyl-imidazole. Process for the preparation of 6α-bromopenicillanic acid 1,1-dioxide according to one of Claims 1 to 11, characterized in that 2-dimethylaminopyridine, 4-dimethylaminopyridine or 1-methylimidazole is used as excipient. Process according to Claim 1, characterized in that R in the formula X is hydrogen and the nitration step with alkyl or cycloalkyl nitrite is carried out without the addition of water and without bromine or hydrogen bromide. Process according to Claim 1, characterized in that R in the formula X is an ester residue and that the nitration step with alkyl or cycloalkyl nitrite is carried out in the presence of a mixture containing bromine, hydrogen bromide and a tertiary amine. Process according to Claim 14, characterized in that the reaction steps are carried out in a non-hydroxyl-containing, inert, organic solvent in the presence of an equimolar amount of a water-soluble carboxylic acid having a pK value of 1.5 to 5.0 in water. and II 80038 Process according to Claim 1, 14 or 15, characterized in that R in the general formula X is an alkylcarbonyloxymethylene residue optionally substituted in the methylene group by one or two methyl groups, and the alkyl group contains 1 to 5 carbon atoms, preferably pivaloyloxymethyl. A 6-diazo-penicillanic acid-1,1-dioxide derivative of the general formula O 0 \ / H S CH2X! / \ / N N C - CC. X I I I, C - N - CH s CH 3 / O 'CO-OR wherein X is hydrogen, chlorine or acetoxy and R is hydrogen or an ester radical, or when R is hydrogen, a salt thereof. Derivative according to Claim 17, characterized in that X is hydrogen. Derivative according to Claim 17, characterized in that X is hydrogen and R is an alkylcarbonyloxymethylene residue, the methylene group of which is optionally substituted by one or two methyl groups and the alkyl group of which contains 1 to 5 carbon atoms, preferably pivaloyloxymethyl. Derivative according to Claim 17, characterized in that X is hydrogen and R is hydrogen or pivaloyloxymethyl, or, when R is hydrogen, a salt thereof. 74 8 0 0 3 8 Derivative according to Claim 17, characterized in that it is 6-diazopenicillanic acid 1,1-dioxide or a salt thereof, preferably a sodium or potassium salt. Derivative according to Claim 17, characterized in that it is pivaloyloxymethyl-6-diazopenicillanate 1,1-dioxide.